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 @var{args}
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. For example:
1912 alias -- Define a new command that is an alias of an existing command
1913 aliases -- Aliases of other commands
1914 d -- Delete some breakpoints or auto-display expressions
1915 del -- Delete some breakpoints or auto-display expressions
1916 delete -- Delete some breakpoints or auto-display expressions
1921 @item complete @var{args}
1922 The @code{complete @var{args}} command lists all the possible completions
1923 for the beginning of a command. Use @var{args} to specify the beginning of the
1924 command you want completed. For example:
1930 @noindent results in:
1941 @noindent This is intended for use by @sc{gnu} Emacs.
1944 In addition to @code{help}, you can use the @value{GDBN} commands @code{info}
1945 and @code{show} to inquire about the state of your program, or the state
1946 of @value{GDBN} itself. Each command supports many topics of inquiry; this
1947 manual introduces each of them in the appropriate context. The listings
1948 under @code{info} and under @code{show} in the Command, Variable, and
1949 Function Index point to all the sub-commands. @xref{Command and Variable
1955 @kindex i @r{(@code{info})}
1957 This command (abbreviated @code{i}) is for describing the state of your
1958 program. For example, you can show the arguments passed to a function
1959 with @code{info args}, list the registers currently in use with @code{info
1960 registers}, or list the breakpoints you have set with @code{info breakpoints}.
1961 You can get a complete list of the @code{info} sub-commands with
1962 @w{@code{help info}}.
1966 You can assign the result of an expression to an environment variable with
1967 @code{set}. For example, you can set the @value{GDBN} prompt to a $-sign with
1968 @code{set prompt $}.
1972 In contrast to @code{info}, @code{show} is for describing the state of
1973 @value{GDBN} itself.
1974 You can change most of the things you can @code{show}, by using the
1975 related command @code{set}; for example, you can control what number
1976 system is used for displays with @code{set radix}, or simply inquire
1977 which is currently in use with @code{show radix}.
1980 To display all the settable parameters and their current
1981 values, you can use @code{show} with no arguments; you may also use
1982 @code{info set}. Both commands produce the same display.
1983 @c FIXME: "info set" violates the rule that "info" is for state of
1984 @c FIXME...program. Ck w/ GNU: "info set" to be called something else,
1985 @c FIXME...or change desc of rule---eg "state of prog and debugging session"?
1989 Here are several miscellaneous @code{show} subcommands, all of which are
1990 exceptional in lacking corresponding @code{set} commands:
1993 @kindex show version
1994 @cindex @value{GDBN} version number
1996 Show what version of @value{GDBN} is running. You should include this
1997 information in @value{GDBN} bug-reports. If multiple versions of
1998 @value{GDBN} are in use at your site, you may need to determine which
1999 version of @value{GDBN} you are running; as @value{GDBN} evolves, new
2000 commands are introduced, and old ones may wither away. Also, many
2001 system vendors ship variant versions of @value{GDBN}, and there are
2002 variant versions of @value{GDBN} in @sc{gnu}/Linux distributions as well.
2003 The version number is the same as the one announced when you start
2006 @kindex show copying
2007 @kindex info copying
2008 @cindex display @value{GDBN} copyright
2011 Display information about permission for copying @value{GDBN}.
2013 @kindex show warranty
2014 @kindex info warranty
2016 @itemx info warranty
2017 Display the @sc{gnu} ``NO WARRANTY'' statement, or a warranty,
2018 if your version of @value{GDBN} comes with one.
2020 @kindex show configuration
2021 @item show configuration
2022 Display detailed information about the way @value{GDBN} was configured
2023 when it was built. This displays the optional arguments passed to the
2024 @file{configure} script and also configuration parameters detected
2025 automatically by @command{configure}. When reporting a @value{GDBN}
2026 bug (@pxref{GDB Bugs}), it is important to include this information in
2032 @chapter Running Programs Under @value{GDBN}
2034 When you run a program under @value{GDBN}, you must first generate
2035 debugging information when you compile it.
2037 You may start @value{GDBN} with its arguments, if any, in an environment
2038 of your choice. If you are doing native debugging, you may redirect
2039 your program's input and output, debug an already running process, or
2040 kill a child process.
2043 * Compilation:: Compiling for debugging
2044 * Starting:: Starting your program
2045 * Arguments:: Your program's arguments
2046 * Environment:: Your program's environment
2048 * Working Directory:: Your program's working directory
2049 * Input/Output:: Your program's input and output
2050 * Attach:: Debugging an already-running process
2051 * Kill Process:: Killing the child process
2053 * Inferiors and Programs:: Debugging multiple inferiors and programs
2054 * Threads:: Debugging programs with multiple threads
2055 * Forks:: Debugging forks
2056 * Checkpoint/Restart:: Setting a @emph{bookmark} to return to later
2060 @section Compiling for Debugging
2062 In order to debug a program effectively, you need to generate
2063 debugging information when you compile it. This debugging information
2064 is stored in the object file; it describes the data type of each
2065 variable or function and the correspondence between source line numbers
2066 and addresses in the executable code.
2068 To request debugging information, specify the @samp{-g} option when you run
2071 Programs that are to be shipped to your customers are compiled with
2072 optimizations, using the @samp{-O} compiler option. However, some
2073 compilers are unable to handle the @samp{-g} and @samp{-O} options
2074 together. Using those compilers, you cannot generate optimized
2075 executables containing debugging information.
2077 @value{NGCC}, the @sc{gnu} C/C@t{++} compiler, supports @samp{-g} with or
2078 without @samp{-O}, making it possible to debug optimized code. We
2079 recommend that you @emph{always} use @samp{-g} whenever you compile a
2080 program. You may think your program is correct, but there is no sense
2081 in pushing your luck. For more information, see @ref{Optimized Code}.
2083 Older versions of the @sc{gnu} C compiler permitted a variant option
2084 @w{@samp{-gg}} for debugging information. @value{GDBN} no longer supports this
2085 format; if your @sc{gnu} C compiler has this option, do not use it.
2087 @value{GDBN} knows about preprocessor macros and can show you their
2088 expansion (@pxref{Macros}). Most compilers do not include information
2089 about preprocessor macros in the debugging information if you specify
2090 the @option{-g} flag alone. Version 3.1 and later of @value{NGCC},
2091 the @sc{gnu} C compiler, provides macro information if you are using
2092 the DWARF debugging format, and specify the option @option{-g3}.
2094 @xref{Debugging Options,,Options for Debugging Your Program or GCC,
2095 gcc, Using the @sc{gnu} Compiler Collection (GCC)}, for more
2096 information on @value{NGCC} options affecting debug information.
2098 You will have the best debugging experience if you use the latest
2099 version of the DWARF debugging format that your compiler supports.
2100 DWARF is currently the most expressive and best supported debugging
2101 format in @value{GDBN}.
2105 @section Starting your Program
2111 @kindex r @r{(@code{run})}
2114 Use the @code{run} command to start your program under @value{GDBN}.
2115 You must first specify the program name with an argument to
2116 @value{GDBN} (@pxref{Invocation, ,Getting In and Out of
2117 @value{GDBN}}), or by using the @code{file} or @code{exec-file}
2118 command (@pxref{Files, ,Commands to Specify Files}).
2122 If you are running your program in an execution environment that
2123 supports processes, @code{run} creates an inferior process and makes
2124 that process run your program. In some environments without processes,
2125 @code{run} jumps to the start of your program. Other targets,
2126 like @samp{remote}, are always running. If you get an error
2127 message like this one:
2130 The "remote" target does not support "run".
2131 Try "help target" or "continue".
2135 then use @code{continue} to run your program. You may need @code{load}
2136 first (@pxref{load}).
2138 The execution of a program is affected by certain information it
2139 receives from its superior. @value{GDBN} provides ways to specify this
2140 information, which you must do @emph{before} starting your program. (You
2141 can change it after starting your program, but such changes only affect
2142 your program the next time you start it.) This information may be
2143 divided into four categories:
2146 @item The @emph{arguments.}
2147 Specify the arguments to give your program as the arguments of the
2148 @code{run} command. If a shell is available on your target, the shell
2149 is used to pass the arguments, so that you may use normal conventions
2150 (such as wildcard expansion or variable substitution) in describing
2152 In Unix systems, you can control which shell is used with the
2153 @code{SHELL} environment variable. If you do not define @code{SHELL},
2154 @value{GDBN} uses the default shell (@file{/bin/sh}). You can disable
2155 use of any shell with the @code{set startup-with-shell} command (see
2158 @item The @emph{environment.}
2159 Your program normally inherits its environment from @value{GDBN}, but you can
2160 use the @value{GDBN} commands @code{set environment} and @code{unset
2161 environment} to change parts of the environment that affect
2162 your program. @xref{Environment, ,Your Program's Environment}.
2164 @item The @emph{working directory.}
2165 You can set your program's working directory with the command
2166 @kbd{set cwd}. If you do not set any working directory with this
2167 command, your program will inherit @value{GDBN}'s working directory if
2168 native debugging, or the remote server's working directory if remote
2169 debugging. @xref{Working Directory, ,Your Program's Working
2172 @item The @emph{standard input and output.}
2173 Your program normally uses the same device for standard input and
2174 standard output as @value{GDBN} is using. You can redirect input and output
2175 in the @code{run} command line, or you can use the @code{tty} command to
2176 set a different device for your program.
2177 @xref{Input/Output, ,Your Program's Input and Output}.
2180 @emph{Warning:} While input and output redirection work, you cannot use
2181 pipes to pass the output of the program you are debugging to another
2182 program; if you attempt this, @value{GDBN} is likely to wind up debugging the
2186 When you issue the @code{run} command, your program begins to execute
2187 immediately. @xref{Stopping, ,Stopping and Continuing}, for discussion
2188 of how to arrange for your program to stop. Once your program has
2189 stopped, you may call functions in your program, using the @code{print}
2190 or @code{call} commands. @xref{Data, ,Examining Data}.
2192 If the modification time of your symbol file has changed since the last
2193 time @value{GDBN} read its symbols, @value{GDBN} discards its symbol
2194 table, and reads it again. When it does this, @value{GDBN} tries to retain
2195 your current breakpoints.
2200 @cindex run to main procedure
2201 The name of the main procedure can vary from language to language.
2202 With C or C@t{++}, the main procedure name is always @code{main}, but
2203 other languages such as Ada do not require a specific name for their
2204 main procedure. The debugger provides a convenient way to start the
2205 execution of the program and to stop at the beginning of the main
2206 procedure, depending on the language used.
2208 The @samp{start} command does the equivalent of setting a temporary
2209 breakpoint at the beginning of the main procedure and then invoking
2210 the @samp{run} command.
2212 @cindex elaboration phase
2213 Some programs contain an @dfn{elaboration} phase where some startup code is
2214 executed before the main procedure is called. This depends on the
2215 languages used to write your program. In C@t{++}, for instance,
2216 constructors for static and global objects are executed before
2217 @code{main} is called. It is therefore possible that the debugger stops
2218 before reaching the main procedure. However, the temporary breakpoint
2219 will remain to halt execution.
2221 Specify the arguments to give to your program as arguments to the
2222 @samp{start} command. These arguments will be given verbatim to the
2223 underlying @samp{run} command. Note that the same arguments will be
2224 reused if no argument is provided during subsequent calls to
2225 @samp{start} or @samp{run}.
2227 It is sometimes necessary to debug the program during elaboration. In
2228 these cases, using the @code{start} command would stop the execution
2229 of your program too late, as the program would have already completed
2230 the elaboration phase. Under these circumstances, either insert
2231 breakpoints in your elaboration code before running your program or
2232 use the @code{starti} command.
2236 @cindex run to first instruction
2237 The @samp{starti} command does the equivalent of setting a temporary
2238 breakpoint at the first instruction of a program's execution and then
2239 invoking the @samp{run} command. For programs containing an
2240 elaboration phase, the @code{starti} command will stop execution at
2241 the start of the elaboration phase.
2243 @anchor{set exec-wrapper}
2244 @kindex set exec-wrapper
2245 @item set exec-wrapper @var{wrapper}
2246 @itemx show exec-wrapper
2247 @itemx unset exec-wrapper
2248 When @samp{exec-wrapper} is set, the specified wrapper is used to
2249 launch programs for debugging. @value{GDBN} starts your program
2250 with a shell command of the form @kbd{exec @var{wrapper}
2251 @var{program}}. Quoting is added to @var{program} and its
2252 arguments, but not to @var{wrapper}, so you should add quotes if
2253 appropriate for your shell. The wrapper runs until it executes
2254 your program, and then @value{GDBN} takes control.
2256 You can use any program that eventually calls @code{execve} with
2257 its arguments as a wrapper. Several standard Unix utilities do
2258 this, e.g.@: @code{env} and @code{nohup}. Any Unix shell script ending
2259 with @code{exec "$@@"} will also work.
2261 For example, you can use @code{env} to pass an environment variable to
2262 the debugged program, without setting the variable in your shell's
2266 (@value{GDBP}) set exec-wrapper env 'LD_PRELOAD=libtest.so'
2270 This command is available when debugging locally on most targets, excluding
2271 @sc{djgpp}, Cygwin, MS Windows, and QNX Neutrino.
2273 @kindex set startup-with-shell
2274 @anchor{set startup-with-shell}
2275 @item set startup-with-shell
2276 @itemx set startup-with-shell on
2277 @itemx set startup-with-shell off
2278 @itemx show startup-with-shell
2279 On Unix systems, by default, if a shell is available on your target,
2280 @value{GDBN}) uses it to start your program. Arguments of the
2281 @code{run} command are passed to the shell, which does variable
2282 substitution, expands wildcard characters and performs redirection of
2283 I/O. In some circumstances, it may be useful to disable such use of a
2284 shell, for example, when debugging the shell itself or diagnosing
2285 startup failures such as:
2289 Starting program: ./a.out
2290 During startup program terminated with signal SIGSEGV, Segmentation fault.
2294 which indicates the shell or the wrapper specified with
2295 @samp{exec-wrapper} crashed, not your program. Most often, this is
2296 caused by something odd in your shell's non-interactive mode
2297 initialization file---such as @file{.cshrc} for C-shell,
2298 $@file{.zshenv} for the Z shell, or the file specified in the
2299 @samp{BASH_ENV} environment variable for BASH.
2301 @anchor{set auto-connect-native-target}
2302 @kindex set auto-connect-native-target
2303 @item set auto-connect-native-target
2304 @itemx set auto-connect-native-target on
2305 @itemx set auto-connect-native-target off
2306 @itemx show auto-connect-native-target
2308 By default, if not connected to any target yet (e.g., with
2309 @code{target remote}), the @code{run} command starts your program as a
2310 native process under @value{GDBN}, on your local machine. If you're
2311 sure you don't want to debug programs on your local machine, you can
2312 tell @value{GDBN} to not connect to the native target automatically
2313 with the @code{set auto-connect-native-target off} command.
2315 If @code{on}, which is the default, and if @value{GDBN} is not
2316 connected to a target already, the @code{run} command automaticaly
2317 connects to the native target, if one is available.
2319 If @code{off}, and if @value{GDBN} is not connected to a target
2320 already, the @code{run} command fails with an error:
2324 Don't know how to run. Try "help target".
2327 If @value{GDBN} is already connected to a target, @value{GDBN} always
2328 uses it with the @code{run} command.
2330 In any case, you can explicitly connect to the native target with the
2331 @code{target native} command. For example,
2334 (@value{GDBP}) set auto-connect-native-target off
2336 Don't know how to run. Try "help target".
2337 (@value{GDBP}) target native
2339 Starting program: ./a.out
2340 [Inferior 1 (process 10421) exited normally]
2343 In case you connected explicitly to the @code{native} target,
2344 @value{GDBN} remains connected even if all inferiors exit, ready for
2345 the next @code{run} command. Use the @code{disconnect} command to
2348 Examples of other commands that likewise respect the
2349 @code{auto-connect-native-target} setting: @code{attach}, @code{info
2350 proc}, @code{info os}.
2352 @kindex set disable-randomization
2353 @item set disable-randomization
2354 @itemx set disable-randomization on
2355 This option (enabled by default in @value{GDBN}) will turn off the native
2356 randomization of the virtual address space of the started program. This option
2357 is useful for multiple debugging sessions to make the execution better
2358 reproducible and memory addresses reusable across debugging sessions.
2360 This feature is implemented only on certain targets, including @sc{gnu}/Linux.
2361 On @sc{gnu}/Linux you can get the same behavior using
2364 (@value{GDBP}) set exec-wrapper setarch `uname -m` -R
2367 @item set disable-randomization off
2368 Leave the behavior of the started executable unchanged. Some bugs rear their
2369 ugly heads only when the program is loaded at certain addresses. If your bug
2370 disappears when you run the program under @value{GDBN}, that might be because
2371 @value{GDBN} by default disables the address randomization on platforms, such
2372 as @sc{gnu}/Linux, which do that for stand-alone programs. Use @kbd{set
2373 disable-randomization off} to try to reproduce such elusive bugs.
2375 On targets where it is available, virtual address space randomization
2376 protects the programs against certain kinds of security attacks. In these
2377 cases the attacker needs to know the exact location of a concrete executable
2378 code. Randomizing its location makes it impossible to inject jumps misusing
2379 a code at its expected addresses.
2381 Prelinking shared libraries provides a startup performance advantage but it
2382 makes addresses in these libraries predictable for privileged processes by
2383 having just unprivileged access at the target system. Reading the shared
2384 library binary gives enough information for assembling the malicious code
2385 misusing it. Still even a prelinked shared library can get loaded at a new
2386 random address just requiring the regular relocation process during the
2387 startup. Shared libraries not already prelinked are always loaded at
2388 a randomly chosen address.
2390 Position independent executables (PIE) contain position independent code
2391 similar to the shared libraries and therefore such executables get loaded at
2392 a randomly chosen address upon startup. PIE executables always load even
2393 already prelinked shared libraries at a random address. You can build such
2394 executable using @command{gcc -fPIE -pie}.
2396 Heap (malloc storage), stack and custom mmap areas are always placed randomly
2397 (as long as the randomization is enabled).
2399 @item show disable-randomization
2400 Show the current setting of the explicit disable of the native randomization of
2401 the virtual address space of the started program.
2406 @section Your Program's Arguments
2408 @cindex arguments (to your program)
2409 The arguments to your program can be specified by the arguments of the
2411 They are passed to a shell, which expands wildcard characters and
2412 performs redirection of I/O, and thence to your program. Your
2413 @code{SHELL} environment variable (if it exists) specifies what shell
2414 @value{GDBN} uses. If you do not define @code{SHELL}, @value{GDBN} uses
2415 the default shell (@file{/bin/sh} on Unix).
2417 On non-Unix systems, the program is usually invoked directly by
2418 @value{GDBN}, which emulates I/O redirection via the appropriate system
2419 calls, and the wildcard characters are expanded by the startup code of
2420 the program, not by the shell.
2422 @code{run} with no arguments uses the same arguments used by the previous
2423 @code{run}, or those set by the @code{set args} command.
2428 Specify the arguments to be used the next time your program is run. If
2429 @code{set args} has no arguments, @code{run} executes your program
2430 with no arguments. Once you have run your program with arguments,
2431 using @code{set args} before the next @code{run} is the only way to run
2432 it again without arguments.
2436 Show the arguments to give your program when it is started.
2440 @section Your Program's Environment
2442 @cindex environment (of your program)
2443 The @dfn{environment} consists of a set of environment variables and
2444 their values. Environment variables conventionally record such things as
2445 your user name, your home directory, your terminal type, and your search
2446 path for programs to run. Usually you set up environment variables with
2447 the shell and they are inherited by all the other programs you run. When
2448 debugging, it can be useful to try running your program with a modified
2449 environment without having to start @value{GDBN} over again.
2453 @item path @var{directory}
2454 Add @var{directory} to the front of the @code{PATH} environment variable
2455 (the search path for executables) that will be passed to your program.
2456 The value of @code{PATH} used by @value{GDBN} does not change.
2457 You may specify several directory names, separated by whitespace or by a
2458 system-dependent separator character (@samp{:} on Unix, @samp{;} on
2459 MS-DOS and MS-Windows). If @var{directory} is already in the path, it
2460 is moved to the front, so it is searched sooner.
2462 You can use the string @samp{$cwd} to refer to whatever is the current
2463 working directory at the time @value{GDBN} searches the path. If you
2464 use @samp{.} instead, it refers to the directory where you executed the
2465 @code{path} command. @value{GDBN} replaces @samp{.} in the
2466 @var{directory} argument (with the current path) before adding
2467 @var{directory} to the search path.
2468 @c 'path' is explicitly nonrepeatable, but RMS points out it is silly to
2469 @c document that, since repeating it would be a no-op.
2473 Display the list of search paths for executables (the @code{PATH}
2474 environment variable).
2476 @kindex show environment
2477 @item show environment @r{[}@var{varname}@r{]}
2478 Print the value of environment variable @var{varname} to be given to
2479 your program when it starts. If you do not supply @var{varname},
2480 print the names and values of all environment variables to be given to
2481 your program. You can abbreviate @code{environment} as @code{env}.
2483 @kindex set environment
2484 @anchor{set environment}
2485 @item set environment @var{varname} @r{[}=@var{value}@r{]}
2486 Set environment variable @var{varname} to @var{value}. The value
2487 changes for your program (and the shell @value{GDBN} uses to launch
2488 it), not for @value{GDBN} itself. The @var{value} may be any string; the
2489 values of environment variables are just strings, and any
2490 interpretation is supplied by your program itself. The @var{value}
2491 parameter is optional; if it is eliminated, the variable is set to a
2493 @c "any string" here does not include leading, trailing
2494 @c blanks. Gnu asks: does anyone care?
2496 For example, this command:
2503 tells the debugged program, when subsequently run, that its user is named
2504 @samp{foo}. (The spaces around @samp{=} are used for clarity here; they
2505 are not actually required.)
2507 Note that on Unix systems, @value{GDBN} runs your program via a shell,
2508 which also inherits the environment set with @code{set environment}.
2509 If necessary, you can avoid that by using the @samp{env} program as a
2510 wrapper instead of using @code{set environment}. @xref{set
2511 exec-wrapper}, for an example doing just that.
2513 Environment variables that are set by the user are also transmitted to
2514 @command{gdbserver} to be used when starting the remote inferior.
2515 @pxref{QEnvironmentHexEncoded}.
2517 @kindex unset environment
2518 @anchor{unset environment}
2519 @item unset environment @var{varname}
2520 Remove variable @var{varname} from the environment to be passed to your
2521 program. This is different from @samp{set env @var{varname} =};
2522 @code{unset environment} removes the variable from the environment,
2523 rather than assigning it an empty value.
2525 Environment variables that are unset by the user are also unset on
2526 @command{gdbserver} when starting the remote inferior.
2527 @pxref{QEnvironmentUnset}.
2530 @emph{Warning:} On Unix systems, @value{GDBN} runs your program using
2531 the shell indicated by your @code{SHELL} environment variable if it
2532 exists (or @code{/bin/sh} if not). If your @code{SHELL} variable
2533 names a shell that runs an initialization file when started
2534 non-interactively---such as @file{.cshrc} for C-shell, $@file{.zshenv}
2535 for the Z shell, or the file specified in the @samp{BASH_ENV}
2536 environment variable for BASH---any variables you set in that file
2537 affect your program. You may wish to move setting of environment
2538 variables to files that are only run when you sign on, such as
2539 @file{.login} or @file{.profile}.
2541 @node Working Directory
2542 @section Your Program's Working Directory
2544 @cindex working directory (of your program)
2545 Each time you start your program with @code{run}, the inferior will be
2546 initialized with the current working directory specified by the
2547 @kbd{set cwd} command. If no directory has been specified by this
2548 command, then the inferior will inherit @value{GDBN}'s current working
2549 directory as its working directory if native debugging, or it will
2550 inherit the remote server's current working directory if remote
2555 @cindex change inferior's working directory
2556 @anchor{set cwd command}
2557 @item set cwd @r{[}@var{directory}@r{]}
2558 Set the inferior's working directory to @var{directory}, which will be
2559 @code{glob}-expanded in order to resolve tildes (@file{~}). If no
2560 argument has been specified, the command clears the setting and resets
2561 it to an empty state. This setting has no effect on @value{GDBN}'s
2562 working directory, and it only takes effect the next time you start
2563 the inferior. The @file{~} in @var{directory} is a short for the
2564 @dfn{home directory}, usually pointed to by the @env{HOME} environment
2565 variable. On MS-Windows, if @env{HOME} is not defined, @value{GDBN}
2566 uses the concatenation of @env{HOMEDRIVE} and @env{HOMEPATH} as
2569 You can also change @value{GDBN}'s current working directory by using
2570 the @code{cd} command.
2574 @cindex show inferior's working directory
2576 Show the inferior's working directory. If no directory has been
2577 specified by @kbd{set cwd}, then the default inferior's working
2578 directory is the same as @value{GDBN}'s working directory.
2581 @cindex change @value{GDBN}'s working directory
2583 @item cd @r{[}@var{directory}@r{]}
2584 Set the @value{GDBN} working directory to @var{directory}. If not
2585 given, @var{directory} uses @file{'~'}.
2587 The @value{GDBN} working directory serves as a default for the
2588 commands that specify files for @value{GDBN} to operate on.
2589 @xref{Files, ,Commands to Specify Files}.
2590 @xref{set cwd command}.
2594 Print the @value{GDBN} working directory.
2597 It is generally impossible to find the current working directory of
2598 the process being debugged (since a program can change its directory
2599 during its run). If you work on a system where @value{GDBN} supports
2600 the @code{info proc} command (@pxref{Process Information}), you can
2601 use the @code{info proc} command to find out the
2602 current working directory of the debuggee.
2605 @section Your Program's Input and Output
2610 By default, the program you run under @value{GDBN} does input and output to
2611 the same terminal that @value{GDBN} uses. @value{GDBN} switches the terminal
2612 to its own terminal modes to interact with you, but it records the terminal
2613 modes your program was using and switches back to them when you continue
2614 running your program.
2617 @kindex info terminal
2619 Displays information recorded by @value{GDBN} about the terminal modes your
2623 You can redirect your program's input and/or output using shell
2624 redirection with the @code{run} command. For example,
2631 starts your program, diverting its output to the file @file{outfile}.
2634 @cindex controlling terminal
2635 Another way to specify where your program should do input and output is
2636 with the @code{tty} command. This command accepts a file name as
2637 argument, and causes this file to be the default for future @code{run}
2638 commands. It also resets the controlling terminal for the child
2639 process, for future @code{run} commands. For example,
2646 directs that processes started with subsequent @code{run} commands
2647 default to do input and output on the terminal @file{/dev/ttyb} and have
2648 that as their controlling terminal.
2650 An explicit redirection in @code{run} overrides the @code{tty} command's
2651 effect on the input/output device, but not its effect on the controlling
2654 When you use the @code{tty} command or redirect input in the @code{run}
2655 command, only the input @emph{for your program} is affected. The input
2656 for @value{GDBN} still comes from your terminal. @code{tty} is an alias
2657 for @code{set inferior-tty}.
2659 @cindex inferior tty
2660 @cindex set inferior controlling terminal
2661 You can use the @code{show inferior-tty} command to tell @value{GDBN} to
2662 display the name of the terminal that will be used for future runs of your
2666 @item set inferior-tty [ @var{tty} ]
2667 @kindex set inferior-tty
2668 Set the tty for the program being debugged to @var{tty}. Omitting @var{tty}
2669 restores the default behavior, which is to use the same terminal as
2672 @item show inferior-tty
2673 @kindex show inferior-tty
2674 Show the current tty for the program being debugged.
2678 @section Debugging an Already-running Process
2683 @item attach @var{process-id}
2684 This command attaches to a running process---one that was started
2685 outside @value{GDBN}. (@code{info files} shows your active
2686 targets.) The command takes as argument a process ID. The usual way to
2687 find out the @var{process-id} of a Unix process is with the @code{ps} utility,
2688 or with the @samp{jobs -l} shell command.
2690 @code{attach} does not repeat if you press @key{RET} a second time after
2691 executing the command.
2694 To use @code{attach}, your program must be running in an environment
2695 which supports processes; for example, @code{attach} does not work for
2696 programs on bare-board targets that lack an operating system. You must
2697 also have permission to send the process a signal.
2699 When you use @code{attach}, the debugger finds the program running in
2700 the process first by looking in the current working directory, then (if
2701 the program is not found) by using the source file search path
2702 (@pxref{Source Path, ,Specifying Source Directories}). You can also use
2703 the @code{file} command to load the program. @xref{Files, ,Commands to
2706 The first thing @value{GDBN} does after arranging to debug the specified
2707 process is to stop it. You can examine and modify an attached process
2708 with all the @value{GDBN} commands that are ordinarily available when
2709 you start processes with @code{run}. You can insert breakpoints; you
2710 can step and continue; you can modify storage. If you would rather the
2711 process continue running, you may use the @code{continue} command after
2712 attaching @value{GDBN} to the process.
2717 When you have finished debugging the attached process, you can use the
2718 @code{detach} command to release it from @value{GDBN} control. Detaching
2719 the process continues its execution. After the @code{detach} command,
2720 that process and @value{GDBN} become completely independent once more, and you
2721 are ready to @code{attach} another process or start one with @code{run}.
2722 @code{detach} does not repeat if you press @key{RET} again after
2723 executing the command.
2726 If you exit @value{GDBN} while you have an attached process, you detach
2727 that process. If you use the @code{run} command, you kill that process.
2728 By default, @value{GDBN} asks for confirmation if you try to do either of these
2729 things; you can control whether or not you need to confirm by using the
2730 @code{set confirm} command (@pxref{Messages/Warnings, ,Optional Warnings and
2734 @section Killing the Child Process
2739 Kill the child process in which your program is running under @value{GDBN}.
2742 This command is useful if you wish to debug a core dump instead of a
2743 running process. @value{GDBN} ignores any core dump file while your program
2746 On some operating systems, a program cannot be executed outside @value{GDBN}
2747 while you have breakpoints set on it inside @value{GDBN}. You can use the
2748 @code{kill} command in this situation to permit running your program
2749 outside the debugger.
2751 The @code{kill} command is also useful if you wish to recompile and
2752 relink your program, since on many systems it is impossible to modify an
2753 executable file while it is running in a process. In this case, when you
2754 next type @code{run}, @value{GDBN} notices that the file has changed, and
2755 reads the symbol table again (while trying to preserve your current
2756 breakpoint settings).
2758 @node Inferiors and Programs
2759 @section Debugging Multiple Inferiors and Programs
2761 @value{GDBN} lets you run and debug multiple programs in a single
2762 session. In addition, @value{GDBN} on some systems may let you run
2763 several programs simultaneously (otherwise you have to exit from one
2764 before starting another). In the most general case, you can have
2765 multiple threads of execution in each of multiple processes, launched
2766 from multiple executables.
2769 @value{GDBN} represents the state of each program execution with an
2770 object called an @dfn{inferior}. An inferior typically corresponds to
2771 a process, but is more general and applies also to targets that do not
2772 have processes. Inferiors may be created before a process runs, and
2773 may be retained after a process exits. Inferiors have unique
2774 identifiers that are different from process ids. Usually each
2775 inferior will also have its own distinct address space, although some
2776 embedded targets may have several inferiors running in different parts
2777 of a single address space. Each inferior may in turn have multiple
2778 threads running in it.
2780 To find out what inferiors exist at any moment, use @w{@code{info
2784 @kindex info inferiors [ @var{id}@dots{} ]
2785 @item info inferiors
2786 Print a list of all inferiors currently being managed by @value{GDBN}.
2787 By default all inferiors are printed, but the argument @var{id}@dots{}
2788 -- a space separated list of inferior numbers -- can be used to limit
2789 the display to just the requested inferiors.
2791 @value{GDBN} displays for each inferior (in this order):
2795 the inferior number assigned by @value{GDBN}
2798 the target system's inferior identifier
2801 the name of the executable the inferior is running.
2806 An asterisk @samp{*} preceding the @value{GDBN} inferior number
2807 indicates the current inferior.
2811 @c end table here to get a little more width for example
2814 (@value{GDBP}) info inferiors
2815 Num Description Executable
2816 2 process 2307 hello
2817 * 1 process 3401 goodbye
2820 To switch focus between inferiors, use the @code{inferior} command:
2823 @kindex inferior @var{infno}
2824 @item inferior @var{infno}
2825 Make inferior number @var{infno} the current inferior. The argument
2826 @var{infno} is the inferior number assigned by @value{GDBN}, as shown
2827 in the first field of the @samp{info inferiors} display.
2830 @vindex $_inferior@r{, convenience variable}
2831 The debugger convenience variable @samp{$_inferior} contains the
2832 number of the current inferior. You may find this useful in writing
2833 breakpoint conditional expressions, command scripts, and so forth.
2834 @xref{Convenience Vars,, Convenience Variables}, for general
2835 information on convenience variables.
2837 You can get multiple executables into a debugging session via the
2838 @code{add-inferior} and @w{@code{clone-inferior}} commands. On some
2839 systems @value{GDBN} can add inferiors to the debug session
2840 automatically by following calls to @code{fork} and @code{exec}. To
2841 remove inferiors from the debugging session use the
2842 @w{@code{remove-inferiors}} command.
2845 @kindex add-inferior
2846 @item add-inferior [ -copies @var{n} ] [ -exec @var{executable} ]
2847 Adds @var{n} inferiors to be run using @var{executable} as the
2848 executable; @var{n} defaults to 1. If no executable is specified,
2849 the inferiors begins empty, with no program. You can still assign or
2850 change the program assigned to the inferior at any time by using the
2851 @code{file} command with the executable name as its argument.
2853 @kindex clone-inferior
2854 @item clone-inferior [ -copies @var{n} ] [ @var{infno} ]
2855 Adds @var{n} inferiors ready to execute the same program as inferior
2856 @var{infno}; @var{n} defaults to 1, and @var{infno} defaults to the
2857 number of the current inferior. This is a convenient command when you
2858 want to run another instance of the inferior you are debugging.
2861 (@value{GDBP}) info inferiors
2862 Num Description Executable
2863 * 1 process 29964 helloworld
2864 (@value{GDBP}) clone-inferior
2867 (@value{GDBP}) info inferiors
2868 Num Description Executable
2870 * 1 process 29964 helloworld
2873 You can now simply switch focus to inferior 2 and run it.
2875 @kindex remove-inferiors
2876 @item remove-inferiors @var{infno}@dots{}
2877 Removes the inferior or inferiors @var{infno}@dots{}. It is not
2878 possible to remove an inferior that is running with this command. For
2879 those, use the @code{kill} or @code{detach} command first.
2883 To quit debugging one of the running inferiors that is not the current
2884 inferior, you can either detach from it by using the @w{@code{detach
2885 inferior}} command (allowing it to run independently), or kill it
2886 using the @w{@code{kill inferiors}} command:
2889 @kindex detach inferiors @var{infno}@dots{}
2890 @item detach inferior @var{infno}@dots{}
2891 Detach from the inferior or inferiors identified by @value{GDBN}
2892 inferior number(s) @var{infno}@dots{}. Note that the inferior's entry
2893 still stays on the list of inferiors shown by @code{info inferiors},
2894 but its Description will show @samp{<null>}.
2896 @kindex kill inferiors @var{infno}@dots{}
2897 @item kill inferiors @var{infno}@dots{}
2898 Kill the inferior or inferiors identified by @value{GDBN} inferior
2899 number(s) @var{infno}@dots{}. Note that the inferior's entry still
2900 stays on the list of inferiors shown by @code{info inferiors}, but its
2901 Description will show @samp{<null>}.
2904 After the successful completion of a command such as @code{detach},
2905 @code{detach inferiors}, @code{kill} or @code{kill inferiors}, or after
2906 a normal process exit, the inferior is still valid and listed with
2907 @code{info inferiors}, ready to be restarted.
2910 To be notified when inferiors are started or exit under @value{GDBN}'s
2911 control use @w{@code{set print inferior-events}}:
2914 @kindex set print inferior-events
2915 @cindex print messages on inferior start and exit
2916 @item set print inferior-events
2917 @itemx set print inferior-events on
2918 @itemx set print inferior-events off
2919 The @code{set print inferior-events} command allows you to enable or
2920 disable printing of messages when @value{GDBN} notices that new
2921 inferiors have started or that inferiors have exited or have been
2922 detached. By default, these messages will not be printed.
2924 @kindex show print inferior-events
2925 @item show print inferior-events
2926 Show whether messages will be printed when @value{GDBN} detects that
2927 inferiors have started, exited or have been detached.
2930 Many commands will work the same with multiple programs as with a
2931 single program: e.g., @code{print myglobal} will simply display the
2932 value of @code{myglobal} in the current inferior.
2935 Occasionaly, when debugging @value{GDBN} itself, it may be useful to
2936 get more info about the relationship of inferiors, programs, address
2937 spaces in a debug session. You can do that with the @w{@code{maint
2938 info program-spaces}} command.
2941 @kindex maint info program-spaces
2942 @item maint info program-spaces
2943 Print a list of all program spaces currently being managed by
2946 @value{GDBN} displays for each program space (in this order):
2950 the program space number assigned by @value{GDBN}
2953 the name of the executable loaded into the program space, with e.g.,
2954 the @code{file} command.
2959 An asterisk @samp{*} preceding the @value{GDBN} program space number
2960 indicates the current program space.
2962 In addition, below each program space line, @value{GDBN} prints extra
2963 information that isn't suitable to display in tabular form. For
2964 example, the list of inferiors bound to the program space.
2967 (@value{GDBP}) maint info program-spaces
2971 Bound inferiors: ID 1 (process 21561)
2974 Here we can see that no inferior is running the program @code{hello},
2975 while @code{process 21561} is running the program @code{goodbye}. On
2976 some targets, it is possible that multiple inferiors are bound to the
2977 same program space. The most common example is that of debugging both
2978 the parent and child processes of a @code{vfork} call. For example,
2981 (@value{GDBP}) maint info program-spaces
2984 Bound inferiors: ID 2 (process 18050), ID 1 (process 18045)
2987 Here, both inferior 2 and inferior 1 are running in the same program
2988 space as a result of inferior 1 having executed a @code{vfork} call.
2992 @section Debugging Programs with Multiple Threads
2994 @cindex threads of execution
2995 @cindex multiple threads
2996 @cindex switching threads
2997 In some operating systems, such as GNU/Linux and Solaris, a single program
2998 may have more than one @dfn{thread} of execution. The precise semantics
2999 of threads differ from one operating system to another, but in general
3000 the threads of a single program are akin to multiple processes---except
3001 that they share one address space (that is, they can all examine and
3002 modify the same variables). On the other hand, each thread has its own
3003 registers and execution stack, and perhaps private memory.
3005 @value{GDBN} provides these facilities for debugging multi-thread
3009 @item automatic notification of new threads
3010 @item @samp{thread @var{thread-id}}, a command to switch among threads
3011 @item @samp{info threads}, a command to inquire about existing threads
3012 @item @samp{thread apply [@var{thread-id-list} | all] @var{args}},
3013 a command to apply a command to a list of threads
3014 @item thread-specific breakpoints
3015 @item @samp{set print thread-events}, which controls printing of
3016 messages on thread start and exit.
3017 @item @samp{set libthread-db-search-path @var{path}}, which lets
3018 the user specify which @code{libthread_db} to use if the default choice
3019 isn't compatible with the program.
3022 @cindex focus of debugging
3023 @cindex current thread
3024 The @value{GDBN} thread debugging facility allows you to observe all
3025 threads while your program runs---but whenever @value{GDBN} takes
3026 control, one thread in particular is always the focus of debugging.
3027 This thread is called the @dfn{current thread}. Debugging commands show
3028 program information from the perspective of the current thread.
3030 @cindex @code{New} @var{systag} message
3031 @cindex thread identifier (system)
3032 @c FIXME-implementors!! It would be more helpful if the [New...] message
3033 @c included GDB's numeric thread handle, so you could just go to that
3034 @c thread without first checking `info threads'.
3035 Whenever @value{GDBN} detects a new thread in your program, it displays
3036 the target system's identification for the thread with a message in the
3037 form @samp{[New @var{systag}]}, where @var{systag} is a thread identifier
3038 whose form varies depending on the particular system. For example, on
3039 @sc{gnu}/Linux, you might see
3042 [New Thread 0x41e02940 (LWP 25582)]
3046 when @value{GDBN} notices a new thread. In contrast, on other systems,
3047 the @var{systag} is simply something like @samp{process 368}, with no
3050 @c FIXME!! (1) Does the [New...] message appear even for the very first
3051 @c thread of a program, or does it only appear for the
3052 @c second---i.e.@: when it becomes obvious we have a multithread
3054 @c (2) *Is* there necessarily a first thread always? Or do some
3055 @c multithread systems permit starting a program with multiple
3056 @c threads ab initio?
3058 @anchor{thread numbers}
3059 @cindex thread number, per inferior
3060 @cindex thread identifier (GDB)
3061 For debugging purposes, @value{GDBN} associates its own thread number
3062 ---always a single integer---with each thread of an inferior. This
3063 number is unique between all threads of an inferior, but not unique
3064 between threads of different inferiors.
3066 @cindex qualified thread ID
3067 You can refer to a given thread in an inferior using the qualified
3068 @var{inferior-num}.@var{thread-num} syntax, also known as
3069 @dfn{qualified thread ID}, with @var{inferior-num} being the inferior
3070 number and @var{thread-num} being the thread number of the given
3071 inferior. For example, thread @code{2.3} refers to thread number 3 of
3072 inferior 2. If you omit @var{inferior-num} (e.g., @code{thread 3}),
3073 then @value{GDBN} infers you're referring to a thread of the current
3076 Until you create a second inferior, @value{GDBN} does not show the
3077 @var{inferior-num} part of thread IDs, even though you can always use
3078 the full @var{inferior-num}.@var{thread-num} form to refer to threads
3079 of inferior 1, the initial inferior.
3081 @anchor{thread ID lists}
3082 @cindex thread ID lists
3083 Some commands accept a space-separated @dfn{thread ID list} as
3084 argument. A list element can be:
3088 A thread ID as shown in the first field of the @samp{info threads}
3089 display, with or without an inferior qualifier. E.g., @samp{2.1} or
3093 A range of thread numbers, again with or without an inferior
3094 qualifier, as in @var{inf}.@var{thr1}-@var{thr2} or
3095 @var{thr1}-@var{thr2}. E.g., @samp{1.2-4} or @samp{2-4}.
3098 All threads of an inferior, specified with a star wildcard, with or
3099 without an inferior qualifier, as in @var{inf}.@code{*} (e.g.,
3100 @samp{1.*}) or @code{*}. The former refers to all threads of the
3101 given inferior, and the latter form without an inferior qualifier
3102 refers to all threads of the current inferior.
3106 For example, if the current inferior is 1, and inferior 7 has one
3107 thread with ID 7.1, the thread list @samp{1 2-3 4.5 6.7-9 7.*}
3108 includes threads 1 to 3 of inferior 1, thread 5 of inferior 4, threads
3109 7 to 9 of inferior 6 and all threads of inferior 7. That is, in
3110 expanded qualified form, the same as @samp{1.1 1.2 1.3 4.5 6.7 6.8 6.9
3114 @anchor{global thread numbers}
3115 @cindex global thread number
3116 @cindex global thread identifier (GDB)
3117 In addition to a @emph{per-inferior} number, each thread is also
3118 assigned a unique @emph{global} number, also known as @dfn{global
3119 thread ID}, a single integer. Unlike the thread number component of
3120 the thread ID, no two threads have the same global ID, even when
3121 you're debugging multiple inferiors.
3123 From @value{GDBN}'s perspective, a process always has at least one
3124 thread. In other words, @value{GDBN} assigns a thread number to the
3125 program's ``main thread'' even if the program is not multi-threaded.
3127 @vindex $_thread@r{, convenience variable}
3128 @vindex $_gthread@r{, convenience variable}
3129 The debugger convenience variables @samp{$_thread} and
3130 @samp{$_gthread} contain, respectively, the per-inferior thread number
3131 and the global thread number of the current thread. You may find this
3132 useful in writing breakpoint conditional expressions, command scripts,
3133 and so forth. @xref{Convenience Vars,, Convenience Variables}, for
3134 general information on convenience variables.
3136 If @value{GDBN} detects the program is multi-threaded, it augments the
3137 usual message about stopping at a breakpoint with the ID and name of
3138 the thread that hit the breakpoint.
3141 Thread 2 "client" hit Breakpoint 1, send_message () at client.c:68
3144 Likewise when the program receives a signal:
3147 Thread 1 "main" received signal SIGINT, Interrupt.
3151 @kindex info threads
3152 @item info threads @r{[}@var{thread-id-list}@r{]}
3154 Display information about one or more threads. With no arguments
3155 displays information about all threads. You can specify the list of
3156 threads that you want to display using the thread ID list syntax
3157 (@pxref{thread ID lists}).
3159 @value{GDBN} displays for each thread (in this order):
3163 the per-inferior thread number assigned by @value{GDBN}
3166 the global thread number assigned by @value{GDBN}, if the @samp{-gid}
3167 option was specified
3170 the target system's thread identifier (@var{systag})
3173 the thread's name, if one is known. A thread can either be named by
3174 the user (see @code{thread name}, below), or, in some cases, by the
3178 the current stack frame summary for that thread
3182 An asterisk @samp{*} to the left of the @value{GDBN} thread number
3183 indicates the current thread.
3187 @c end table here to get a little more width for example
3190 (@value{GDBP}) info threads
3192 * 1 process 35 thread 13 main (argc=1, argv=0x7ffffff8)
3193 2 process 35 thread 23 0x34e5 in sigpause ()
3194 3 process 35 thread 27 0x34e5 in sigpause ()
3198 If you're debugging multiple inferiors, @value{GDBN} displays thread
3199 IDs using the qualified @var{inferior-num}.@var{thread-num} format.
3200 Otherwise, only @var{thread-num} is shown.
3202 If you specify the @samp{-gid} option, @value{GDBN} displays a column
3203 indicating each thread's global thread ID:
3206 (@value{GDBP}) info threads
3207 Id GId Target Id Frame
3208 1.1 1 process 35 thread 13 main (argc=1, argv=0x7ffffff8)
3209 1.2 3 process 35 thread 23 0x34e5 in sigpause ()
3210 1.3 4 process 35 thread 27 0x34e5 in sigpause ()
3211 * 2.1 2 process 65 thread 1 main (argc=1, argv=0x7ffffff8)
3214 On Solaris, you can display more information about user threads with a
3215 Solaris-specific command:
3218 @item maint info sol-threads
3219 @kindex maint info sol-threads
3220 @cindex thread info (Solaris)
3221 Display info on Solaris user threads.
3225 @kindex thread @var{thread-id}
3226 @item thread @var{thread-id}
3227 Make thread ID @var{thread-id} the current thread. The command
3228 argument @var{thread-id} is the @value{GDBN} thread ID, as shown in
3229 the first field of the @samp{info threads} display, with or without an
3230 inferior qualifier (e.g., @samp{2.1} or @samp{1}).
3232 @value{GDBN} responds by displaying the system identifier of the
3233 thread you selected, and its current stack frame summary:
3236 (@value{GDBP}) thread 2
3237 [Switching to thread 2 (Thread 0xb7fdab70 (LWP 12747))]
3238 #0 some_function (ignore=0x0) at example.c:8
3239 8 printf ("hello\n");
3243 As with the @samp{[New @dots{}]} message, the form of the text after
3244 @samp{Switching to} depends on your system's conventions for identifying
3247 @kindex thread apply
3248 @cindex apply command to several threads
3249 @item thread apply [@var{thread-id-list} | all [-ascending]] [@var{flag}]@dots{} @var{command}
3250 The @code{thread apply} command allows you to apply the named
3251 @var{command} to one or more threads. Specify the threads that you
3252 want affected using the thread ID list syntax (@pxref{thread ID
3253 lists}), or specify @code{all} to apply to all threads. To apply a
3254 command to all threads in descending order, type @kbd{thread apply all
3255 @var{command}}. To apply a command to all threads in ascending order,
3256 type @kbd{thread apply all -ascending @var{command}}.
3258 The @var{flag} arguments control what output to produce and how to handle
3259 errors raised when applying @var{command} to a thread. @var{flag}
3260 must start with a @code{-} directly followed by one letter in
3261 @code{qcs}. If several flags are provided, they must be given
3262 individually, such as @code{-c -q}.
3264 By default, @value{GDBN} displays some thread information before the
3265 output produced by @var{command}, and an error raised during the
3266 execution of a @var{command} will abort @code{thread apply}. The
3267 following flags can be used to fine-tune this behavior:
3271 The flag @code{-c}, which stands for @samp{continue}, causes any
3272 errors in @var{command} to be displayed, and the execution of
3273 @code{thread apply} then continues.
3275 The flag @code{-s}, which stands for @samp{silent}, causes any errors
3276 or empty output produced by a @var{command} to be silently ignored.
3277 That is, the execution continues, but the thread information and errors
3280 The flag @code{-q} (@samp{quiet}) disables printing the thread
3284 Flags @code{-c} and @code{-s} cannot be used together.
3287 @cindex apply command to all threads (ignoring errors and empty output)
3288 @item taas @var{command}
3289 Shortcut for @code{thread apply all -s @var{command}}.
3290 Applies @var{command} on all threads, ignoring errors and empty output.
3293 @cindex apply a command to all frames of all threads (ignoring errors and empty output)
3294 @item tfaas @var{command}
3295 Shortcut for @code{thread apply all -s frame apply all -s @var{command}}.
3296 Applies @var{command} on all frames of all threads, ignoring errors
3297 and empty output. Note that the flag @code{-s} is specified twice:
3298 The first @code{-s} ensures that @code{thread apply} only shows the thread
3299 information of the threads for which @code{frame apply} produces
3300 some output. The second @code{-s} is needed to ensure that @code{frame
3301 apply} shows the frame information of a frame only if the
3302 @var{command} successfully produced some output.
3304 It can for example be used to print a local variable or a function
3305 argument without knowing the thread or frame where this variable or argument
3308 (@value{GDBP}) tfaas p some_local_var_i_do_not_remember_where_it_is
3313 @cindex name a thread
3314 @item thread name [@var{name}]
3315 This command assigns a name to the current thread. If no argument is
3316 given, any existing user-specified name is removed. The thread name
3317 appears in the @samp{info threads} display.
3319 On some systems, such as @sc{gnu}/Linux, @value{GDBN} is able to
3320 determine the name of the thread as given by the OS. On these
3321 systems, a name specified with @samp{thread name} will override the
3322 system-give name, and removing the user-specified name will cause
3323 @value{GDBN} to once again display the system-specified name.
3326 @cindex search for a thread
3327 @item thread find [@var{regexp}]
3328 Search for and display thread ids whose name or @var{systag}
3329 matches the supplied regular expression.
3331 As well as being the complement to the @samp{thread name} command,
3332 this command also allows you to identify a thread by its target
3333 @var{systag}. For instance, on @sc{gnu}/Linux, the target @var{systag}
3337 (@value{GDBN}) thread find 26688
3338 Thread 4 has target id 'Thread 0x41e02940 (LWP 26688)'
3339 (@value{GDBN}) info thread 4
3341 4 Thread 0x41e02940 (LWP 26688) 0x00000031ca6cd372 in select ()
3344 @kindex set print thread-events
3345 @cindex print messages on thread start and exit
3346 @item set print thread-events
3347 @itemx set print thread-events on
3348 @itemx set print thread-events off
3349 The @code{set print thread-events} command allows you to enable or
3350 disable printing of messages when @value{GDBN} notices that new threads have
3351 started or that threads have exited. By default, these messages will
3352 be printed if detection of these events is supported by the target.
3353 Note that these messages cannot be disabled on all targets.
3355 @kindex show print thread-events
3356 @item show print thread-events
3357 Show whether messages will be printed when @value{GDBN} detects that threads
3358 have started and exited.
3361 @xref{Thread Stops,,Stopping and Starting Multi-thread Programs}, for
3362 more information about how @value{GDBN} behaves when you stop and start
3363 programs with multiple threads.
3365 @xref{Set Watchpoints,,Setting Watchpoints}, for information about
3366 watchpoints in programs with multiple threads.
3368 @anchor{set libthread-db-search-path}
3370 @kindex set libthread-db-search-path
3371 @cindex search path for @code{libthread_db}
3372 @item set libthread-db-search-path @r{[}@var{path}@r{]}
3373 If this variable is set, @var{path} is a colon-separated list of
3374 directories @value{GDBN} will use to search for @code{libthread_db}.
3375 If you omit @var{path}, @samp{libthread-db-search-path} will be reset to
3376 its default value (@code{$sdir:$pdir} on @sc{gnu}/Linux and Solaris systems).
3377 Internally, the default value comes from the @code{LIBTHREAD_DB_SEARCH_PATH}
3380 On @sc{gnu}/Linux and Solaris systems, @value{GDBN} uses a ``helper''
3381 @code{libthread_db} library to obtain information about threads in the
3382 inferior process. @value{GDBN} will use @samp{libthread-db-search-path}
3383 to find @code{libthread_db}. @value{GDBN} also consults first if inferior
3384 specific thread debugging library loading is enabled
3385 by @samp{set auto-load libthread-db} (@pxref{libthread_db.so.1 file}).
3387 A special entry @samp{$sdir} for @samp{libthread-db-search-path}
3388 refers to the default system directories that are
3389 normally searched for loading shared libraries. The @samp{$sdir} entry
3390 is the only kind not needing to be enabled by @samp{set auto-load libthread-db}
3391 (@pxref{libthread_db.so.1 file}).
3393 A special entry @samp{$pdir} for @samp{libthread-db-search-path}
3394 refers to the directory from which @code{libpthread}
3395 was loaded in the inferior process.
3397 For any @code{libthread_db} library @value{GDBN} finds in above directories,
3398 @value{GDBN} attempts to initialize it with the current inferior process.
3399 If this initialization fails (which could happen because of a version
3400 mismatch between @code{libthread_db} and @code{libpthread}), @value{GDBN}
3401 will unload @code{libthread_db}, and continue with the next directory.
3402 If none of @code{libthread_db} libraries initialize successfully,
3403 @value{GDBN} will issue a warning and thread debugging will be disabled.
3405 Setting @code{libthread-db-search-path} is currently implemented
3406 only on some platforms.
3408 @kindex show libthread-db-search-path
3409 @item show libthread-db-search-path
3410 Display current libthread_db search path.
3412 @kindex set debug libthread-db
3413 @kindex show debug libthread-db
3414 @cindex debugging @code{libthread_db}
3415 @item set debug libthread-db
3416 @itemx show debug libthread-db
3417 Turns on or off display of @code{libthread_db}-related events.
3418 Use @code{1} to enable, @code{0} to disable.
3422 @section Debugging Forks
3424 @cindex fork, debugging programs which call
3425 @cindex multiple processes
3426 @cindex processes, multiple
3427 On most systems, @value{GDBN} has no special support for debugging
3428 programs which create additional processes using the @code{fork}
3429 function. When a program forks, @value{GDBN} will continue to debug the
3430 parent process and the child process will run unimpeded. If you have
3431 set a breakpoint in any code which the child then executes, the child
3432 will get a @code{SIGTRAP} signal which (unless it catches the signal)
3433 will cause it to terminate.
3435 However, if you want to debug the child process there is a workaround
3436 which isn't too painful. Put a call to @code{sleep} in the code which
3437 the child process executes after the fork. It may be useful to sleep
3438 only if a certain environment variable is set, or a certain file exists,
3439 so that the delay need not occur when you don't want to run @value{GDBN}
3440 on the child. While the child is sleeping, use the @code{ps} program to
3441 get its process ID. Then tell @value{GDBN} (a new invocation of
3442 @value{GDBN} if you are also debugging the parent process) to attach to
3443 the child process (@pxref{Attach}). From that point on you can debug
3444 the child process just like any other process which you attached to.
3446 On some systems, @value{GDBN} provides support for debugging programs
3447 that create additional processes using the @code{fork} or @code{vfork}
3448 functions. On @sc{gnu}/Linux platforms, this feature is supported
3449 with kernel version 2.5.46 and later.
3451 The fork debugging commands are supported in native mode and when
3452 connected to @code{gdbserver} in either @code{target remote} mode or
3453 @code{target extended-remote} mode.
3455 By default, when a program forks, @value{GDBN} will continue to debug
3456 the parent process and the child process will run unimpeded.
3458 If you want to follow the child process instead of the parent process,
3459 use the command @w{@code{set follow-fork-mode}}.
3462 @kindex set follow-fork-mode
3463 @item set follow-fork-mode @var{mode}
3464 Set the debugger response to a program call of @code{fork} or
3465 @code{vfork}. A call to @code{fork} or @code{vfork} creates a new
3466 process. The @var{mode} argument can be:
3470 The original process is debugged after a fork. The child process runs
3471 unimpeded. This is the default.
3474 The new process is debugged after a fork. The parent process runs
3479 @kindex show follow-fork-mode
3480 @item show follow-fork-mode
3481 Display the current debugger response to a @code{fork} or @code{vfork} call.
3484 @cindex debugging multiple processes
3485 On Linux, if you want to debug both the parent and child processes, use the
3486 command @w{@code{set detach-on-fork}}.
3489 @kindex set detach-on-fork
3490 @item set detach-on-fork @var{mode}
3491 Tells gdb whether to detach one of the processes after a fork, or
3492 retain debugger control over them both.
3496 The child process (or parent process, depending on the value of
3497 @code{follow-fork-mode}) will be detached and allowed to run
3498 independently. This is the default.
3501 Both processes will be held under the control of @value{GDBN}.
3502 One process (child or parent, depending on the value of
3503 @code{follow-fork-mode}) is debugged as usual, while the other
3508 @kindex show detach-on-fork
3509 @item show detach-on-fork
3510 Show whether detach-on-fork mode is on/off.
3513 If you choose to set @samp{detach-on-fork} mode off, then @value{GDBN}
3514 will retain control of all forked processes (including nested forks).
3515 You can list the forked processes under the control of @value{GDBN} by
3516 using the @w{@code{info inferiors}} command, and switch from one fork
3517 to another by using the @code{inferior} command (@pxref{Inferiors and
3518 Programs, ,Debugging Multiple Inferiors and Programs}).
3520 To quit debugging one of the forked processes, you can either detach
3521 from it by using the @w{@code{detach inferiors}} command (allowing it
3522 to run independently), or kill it using the @w{@code{kill inferiors}}
3523 command. @xref{Inferiors and Programs, ,Debugging Multiple Inferiors
3526 If you ask to debug a child process and a @code{vfork} is followed by an
3527 @code{exec}, @value{GDBN} executes the new target up to the first
3528 breakpoint in the new target. If you have a breakpoint set on
3529 @code{main} in your original program, the breakpoint will also be set on
3530 the child process's @code{main}.
3532 On some systems, when a child process is spawned by @code{vfork}, you
3533 cannot debug the child or parent until an @code{exec} call completes.
3535 If you issue a @code{run} command to @value{GDBN} after an @code{exec}
3536 call executes, the new target restarts. To restart the parent
3537 process, use the @code{file} command with the parent executable name
3538 as its argument. By default, after an @code{exec} call executes,
3539 @value{GDBN} discards the symbols of the previous executable image.
3540 You can change this behaviour with the @w{@code{set follow-exec-mode}}
3544 @kindex set follow-exec-mode
3545 @item set follow-exec-mode @var{mode}
3547 Set debugger response to a program call of @code{exec}. An
3548 @code{exec} call replaces the program image of a process.
3550 @code{follow-exec-mode} can be:
3554 @value{GDBN} creates a new inferior and rebinds the process to this
3555 new inferior. The program the process was running before the
3556 @code{exec} call can be restarted afterwards by restarting the
3562 (@value{GDBP}) info inferiors
3564 Id Description Executable
3567 process 12020 is executing new program: prog2
3568 Program exited normally.
3569 (@value{GDBP}) info inferiors
3570 Id Description Executable
3576 @value{GDBN} keeps the process bound to the same inferior. The new
3577 executable image replaces the previous executable loaded in the
3578 inferior. Restarting the inferior after the @code{exec} call, with
3579 e.g., the @code{run} command, restarts the executable the process was
3580 running after the @code{exec} call. This is the default mode.
3585 (@value{GDBP}) info inferiors
3586 Id Description Executable
3589 process 12020 is executing new program: prog2
3590 Program exited normally.
3591 (@value{GDBP}) info inferiors
3592 Id Description Executable
3599 @code{follow-exec-mode} is supported in native mode and
3600 @code{target extended-remote} mode.
3602 You can use the @code{catch} command to make @value{GDBN} stop whenever
3603 a @code{fork}, @code{vfork}, or @code{exec} call is made. @xref{Set
3604 Catchpoints, ,Setting Catchpoints}.
3606 @node Checkpoint/Restart
3607 @section Setting a @emph{Bookmark} to Return to Later
3612 @cindex snapshot of a process
3613 @cindex rewind program state
3615 On certain operating systems@footnote{Currently, only
3616 @sc{gnu}/Linux.}, @value{GDBN} is able to save a @dfn{snapshot} of a
3617 program's state, called a @dfn{checkpoint}, and come back to it
3620 Returning to a checkpoint effectively undoes everything that has
3621 happened in the program since the @code{checkpoint} was saved. This
3622 includes changes in memory, registers, and even (within some limits)
3623 system state. Effectively, it is like going back in time to the
3624 moment when the checkpoint was saved.
3626 Thus, if you're stepping thru a program and you think you're
3627 getting close to the point where things go wrong, you can save
3628 a checkpoint. Then, if you accidentally go too far and miss
3629 the critical statement, instead of having to restart your program
3630 from the beginning, you can just go back to the checkpoint and
3631 start again from there.
3633 This can be especially useful if it takes a lot of time or
3634 steps to reach the point where you think the bug occurs.
3636 To use the @code{checkpoint}/@code{restart} method of debugging:
3641 Save a snapshot of the debugged program's current execution state.
3642 The @code{checkpoint} command takes no arguments, but each checkpoint
3643 is assigned a small integer id, similar to a breakpoint id.
3645 @kindex info checkpoints
3646 @item info checkpoints
3647 List the checkpoints that have been saved in the current debugging
3648 session. For each checkpoint, the following information will be
3655 @item Source line, or label
3658 @kindex restart @var{checkpoint-id}
3659 @item restart @var{checkpoint-id}
3660 Restore the program state that was saved as checkpoint number
3661 @var{checkpoint-id}. All program variables, registers, stack frames
3662 etc.@: will be returned to the values that they had when the checkpoint
3663 was saved. In essence, gdb will ``wind back the clock'' to the point
3664 in time when the checkpoint was saved.
3666 Note that breakpoints, @value{GDBN} variables, command history etc.
3667 are not affected by restoring a checkpoint. In general, a checkpoint
3668 only restores things that reside in the program being debugged, not in
3671 @kindex delete checkpoint @var{checkpoint-id}
3672 @item delete checkpoint @var{checkpoint-id}
3673 Delete the previously-saved checkpoint identified by @var{checkpoint-id}.
3677 Returning to a previously saved checkpoint will restore the user state
3678 of the program being debugged, plus a significant subset of the system
3679 (OS) state, including file pointers. It won't ``un-write'' data from
3680 a file, but it will rewind the file pointer to the previous location,
3681 so that the previously written data can be overwritten. For files
3682 opened in read mode, the pointer will also be restored so that the
3683 previously read data can be read again.
3685 Of course, characters that have been sent to a printer (or other
3686 external device) cannot be ``snatched back'', and characters received
3687 from eg.@: a serial device can be removed from internal program buffers,
3688 but they cannot be ``pushed back'' into the serial pipeline, ready to
3689 be received again. Similarly, the actual contents of files that have
3690 been changed cannot be restored (at this time).
3692 However, within those constraints, you actually can ``rewind'' your
3693 program to a previously saved point in time, and begin debugging it
3694 again --- and you can change the course of events so as to debug a
3695 different execution path this time.
3697 @cindex checkpoints and process id
3698 Finally, there is one bit of internal program state that will be
3699 different when you return to a checkpoint --- the program's process
3700 id. Each checkpoint will have a unique process id (or @var{pid}),
3701 and each will be different from the program's original @var{pid}.
3702 If your program has saved a local copy of its process id, this could
3703 potentially pose a problem.
3705 @subsection A Non-obvious Benefit of Using Checkpoints
3707 On some systems such as @sc{gnu}/Linux, address space randomization
3708 is performed on new processes for security reasons. This makes it
3709 difficult or impossible to set a breakpoint, or watchpoint, on an
3710 absolute address if you have to restart the program, since the
3711 absolute location of a symbol will change from one execution to the
3714 A checkpoint, however, is an @emph{identical} copy of a process.
3715 Therefore if you create a checkpoint at (eg.@:) the start of main,
3716 and simply return to that checkpoint instead of restarting the
3717 process, you can avoid the effects of address randomization and
3718 your symbols will all stay in the same place.
3721 @chapter Stopping and Continuing
3723 The principal purposes of using a debugger are so that you can stop your
3724 program before it terminates; or so that, if your program runs into
3725 trouble, you can investigate and find out why.
3727 Inside @value{GDBN}, your program may stop for any of several reasons,
3728 such as a signal, a breakpoint, or reaching a new line after a
3729 @value{GDBN} command such as @code{step}. You may then examine and
3730 change variables, set new breakpoints or remove old ones, and then
3731 continue execution. Usually, the messages shown by @value{GDBN} provide
3732 ample explanation of the status of your program---but you can also
3733 explicitly request this information at any time.
3736 @kindex info program
3738 Display information about the status of your program: whether it is
3739 running or not, what process it is, and why it stopped.
3743 * Breakpoints:: Breakpoints, watchpoints, and catchpoints
3744 * Continuing and Stepping:: Resuming execution
3745 * Skipping Over Functions and Files::
3746 Skipping over functions and files
3748 * Thread Stops:: Stopping and starting multi-thread programs
3752 @section Breakpoints, Watchpoints, and Catchpoints
3755 A @dfn{breakpoint} makes your program stop whenever a certain point in
3756 the program is reached. For each breakpoint, you can add conditions to
3757 control in finer detail whether your program stops. You can set
3758 breakpoints with the @code{break} command and its variants (@pxref{Set
3759 Breaks, ,Setting Breakpoints}), to specify the place where your program
3760 should stop by line number, function name or exact address in the
3763 On some systems, you can set breakpoints in shared libraries before
3764 the executable is run.
3767 @cindex data breakpoints
3768 @cindex memory tracing
3769 @cindex breakpoint on memory address
3770 @cindex breakpoint on variable modification
3771 A @dfn{watchpoint} is a special breakpoint that stops your program
3772 when the value of an expression changes. The expression may be a value
3773 of a variable, or it could involve values of one or more variables
3774 combined by operators, such as @samp{a + b}. This is sometimes called
3775 @dfn{data breakpoints}. You must use a different command to set
3776 watchpoints (@pxref{Set Watchpoints, ,Setting Watchpoints}), but aside
3777 from that, you can manage a watchpoint like any other breakpoint: you
3778 enable, disable, and delete both breakpoints and watchpoints using the
3781 You can arrange to have values from your program displayed automatically
3782 whenever @value{GDBN} stops at a breakpoint. @xref{Auto Display,,
3786 @cindex breakpoint on events
3787 A @dfn{catchpoint} is another special breakpoint that stops your program
3788 when a certain kind of event occurs, such as the throwing of a C@t{++}
3789 exception or the loading of a library. As with watchpoints, you use a
3790 different command to set a catchpoint (@pxref{Set Catchpoints, ,Setting
3791 Catchpoints}), but aside from that, you can manage a catchpoint like any
3792 other breakpoint. (To stop when your program receives a signal, use the
3793 @code{handle} command; see @ref{Signals, ,Signals}.)
3795 @cindex breakpoint numbers
3796 @cindex numbers for breakpoints
3797 @value{GDBN} assigns a number to each breakpoint, watchpoint, or
3798 catchpoint when you create it; these numbers are successive integers
3799 starting with one. In many of the commands for controlling various
3800 features of breakpoints you use the breakpoint number to say which
3801 breakpoint you want to change. Each breakpoint may be @dfn{enabled} or
3802 @dfn{disabled}; if disabled, it has no effect on your program until you
3805 @cindex breakpoint ranges
3806 @cindex breakpoint lists
3807 @cindex ranges of breakpoints
3808 @cindex lists of breakpoints
3809 Some @value{GDBN} commands accept a space-separated list of breakpoints
3810 on which to operate. A list element can be either a single breakpoint number,
3811 like @samp{5}, or a range of such numbers, like @samp{5-7}.
3812 When a breakpoint list is given to a command, all breakpoints in that list
3816 * Set Breaks:: Setting breakpoints
3817 * Set Watchpoints:: Setting watchpoints
3818 * Set Catchpoints:: Setting catchpoints
3819 * Delete Breaks:: Deleting breakpoints
3820 * Disabling:: Disabling breakpoints
3821 * Conditions:: Break conditions
3822 * Break Commands:: Breakpoint command lists
3823 * Dynamic Printf:: Dynamic printf
3824 * Save Breakpoints:: How to save breakpoints in a file
3825 * Static Probe Points:: Listing static probe points
3826 * Error in Breakpoints:: ``Cannot insert breakpoints''
3827 * Breakpoint-related Warnings:: ``Breakpoint address adjusted...''
3831 @subsection Setting Breakpoints
3833 @c FIXME LMB what does GDB do if no code on line of breakpt?
3834 @c consider in particular declaration with/without initialization.
3836 @c FIXME 2 is there stuff on this already? break at fun start, already init?
3839 @kindex b @r{(@code{break})}
3840 @vindex $bpnum@r{, convenience variable}
3841 @cindex latest breakpoint
3842 Breakpoints are set with the @code{break} command (abbreviated
3843 @code{b}). The debugger convenience variable @samp{$bpnum} records the
3844 number of the breakpoint you've set most recently; see @ref{Convenience
3845 Vars,, Convenience Variables}, for a discussion of what you can do with
3846 convenience variables.
3849 @item break @var{location}
3850 Set a breakpoint at the given @var{location}, which can specify a
3851 function name, a line number, or an address of an instruction.
3852 (@xref{Specify Location}, for a list of all the possible ways to
3853 specify a @var{location}.) The breakpoint will stop your program just
3854 before it executes any of the code in the specified @var{location}.
3856 When using source languages that permit overloading of symbols, such as
3857 C@t{++}, a function name may refer to more than one possible place to break.
3858 @xref{Ambiguous Expressions,,Ambiguous Expressions}, for a discussion of
3861 It is also possible to insert a breakpoint that will stop the program
3862 only if a specific thread (@pxref{Thread-Specific Breakpoints})
3863 or a specific task (@pxref{Ada Tasks}) hits that breakpoint.
3866 When called without any arguments, @code{break} sets a breakpoint at
3867 the next instruction to be executed in the selected stack frame
3868 (@pxref{Stack, ,Examining the Stack}). In any selected frame but the
3869 innermost, this makes your program stop as soon as control
3870 returns to that frame. This is similar to the effect of a
3871 @code{finish} command in the frame inside the selected frame---except
3872 that @code{finish} does not leave an active breakpoint. If you use
3873 @code{break} without an argument in the innermost frame, @value{GDBN} stops
3874 the next time it reaches the current location; this may be useful
3877 @value{GDBN} normally ignores breakpoints when it resumes execution, until at
3878 least one instruction has been executed. If it did not do this, you
3879 would be unable to proceed past a breakpoint without first disabling the
3880 breakpoint. This rule applies whether or not the breakpoint already
3881 existed when your program stopped.
3883 @item break @dots{} if @var{cond}
3884 Set a breakpoint with condition @var{cond}; evaluate the expression
3885 @var{cond} each time the breakpoint is reached, and stop only if the
3886 value is nonzero---that is, if @var{cond} evaluates as true.
3887 @samp{@dots{}} stands for one of the possible arguments described
3888 above (or no argument) specifying where to break. @xref{Conditions,
3889 ,Break Conditions}, for more information on breakpoint conditions.
3892 @item tbreak @var{args}
3893 Set a breakpoint enabled only for one stop. The @var{args} are the
3894 same as for the @code{break} command, and the breakpoint is set in the same
3895 way, but the breakpoint is automatically deleted after the first time your
3896 program stops there. @xref{Disabling, ,Disabling Breakpoints}.
3899 @cindex hardware breakpoints
3900 @item hbreak @var{args}
3901 Set a hardware-assisted breakpoint. The @var{args} are the same as for the
3902 @code{break} command and the breakpoint is set in the same way, but the
3903 breakpoint requires hardware support and some target hardware may not
3904 have this support. The main purpose of this is EPROM/ROM code
3905 debugging, so you can set a breakpoint at an instruction without
3906 changing the instruction. This can be used with the new trap-generation
3907 provided by SPARClite DSU and most x86-based targets. These targets
3908 will generate traps when a program accesses some data or instruction
3909 address that is assigned to the debug registers. However the hardware
3910 breakpoint registers can take a limited number of breakpoints. For
3911 example, on the DSU, only two data breakpoints can be set at a time, and
3912 @value{GDBN} will reject this command if more than two are used. Delete
3913 or disable unused hardware breakpoints before setting new ones
3914 (@pxref{Disabling, ,Disabling Breakpoints}).
3915 @xref{Conditions, ,Break Conditions}.
3916 For remote targets, you can restrict the number of hardware
3917 breakpoints @value{GDBN} will use, see @ref{set remote
3918 hardware-breakpoint-limit}.
3921 @item thbreak @var{args}
3922 Set a hardware-assisted breakpoint enabled only for one stop. The @var{args}
3923 are the same as for the @code{hbreak} command and the breakpoint is set in
3924 the same way. However, like the @code{tbreak} command,
3925 the breakpoint is automatically deleted after the
3926 first time your program stops there. Also, like the @code{hbreak}
3927 command, the breakpoint requires hardware support and some target hardware
3928 may not have this support. @xref{Disabling, ,Disabling Breakpoints}.
3929 See also @ref{Conditions, ,Break Conditions}.
3932 @cindex regular expression
3933 @cindex breakpoints at functions matching a regexp
3934 @cindex set breakpoints in many functions
3935 @item rbreak @var{regex}
3936 Set breakpoints on all functions matching the regular expression
3937 @var{regex}. This command sets an unconditional breakpoint on all
3938 matches, printing a list of all breakpoints it set. Once these
3939 breakpoints are set, they are treated just like the breakpoints set with
3940 the @code{break} command. You can delete them, disable them, or make
3941 them conditional the same way as any other breakpoint.
3943 In programs using different languages, @value{GDBN} chooses the syntax
3944 to print the list of all breakpoints it sets according to the
3945 @samp{set language} value: using @samp{set language auto}
3946 (see @ref{Automatically, ,Set Language Automatically}) means to use the
3947 language of the breakpoint's function, other values mean to use
3948 the manually specified language (see @ref{Manually, ,Set Language Manually}).
3950 The syntax of the regular expression is the standard one used with tools
3951 like @file{grep}. Note that this is different from the syntax used by
3952 shells, so for instance @code{foo*} matches all functions that include
3953 an @code{fo} followed by zero or more @code{o}s. There is an implicit
3954 @code{.*} leading and trailing the regular expression you supply, so to
3955 match only functions that begin with @code{foo}, use @code{^foo}.
3957 @cindex non-member C@t{++} functions, set breakpoint in
3958 When debugging C@t{++} programs, @code{rbreak} is useful for setting
3959 breakpoints on overloaded functions that are not members of any special
3962 @cindex set breakpoints on all functions
3963 The @code{rbreak} command can be used to set breakpoints in
3964 @strong{all} the functions in a program, like this:
3967 (@value{GDBP}) rbreak .
3970 @item rbreak @var{file}:@var{regex}
3971 If @code{rbreak} is called with a filename qualification, it limits
3972 the search for functions matching the given regular expression to the
3973 specified @var{file}. This can be used, for example, to set breakpoints on
3974 every function in a given file:
3977 (@value{GDBP}) rbreak file.c:.
3980 The colon separating the filename qualifier from the regex may
3981 optionally be surrounded by spaces.
3983 @kindex info breakpoints
3984 @cindex @code{$_} and @code{info breakpoints}
3985 @item info breakpoints @r{[}@var{list}@dots{}@r{]}
3986 @itemx info break @r{[}@var{list}@dots{}@r{]}
3987 Print a table of all breakpoints, watchpoints, and catchpoints set and
3988 not deleted. Optional argument @var{n} means print information only
3989 about the specified breakpoint(s) (or watchpoint(s) or catchpoint(s)).
3990 For each breakpoint, following columns are printed:
3993 @item Breakpoint Numbers
3995 Breakpoint, watchpoint, or catchpoint.
3997 Whether the breakpoint is marked to be disabled or deleted when hit.
3998 @item Enabled or Disabled
3999 Enabled breakpoints are marked with @samp{y}. @samp{n} marks breakpoints
4000 that are not enabled.
4002 Where the breakpoint is in your program, as a memory address. For a
4003 pending breakpoint whose address is not yet known, this field will
4004 contain @samp{<PENDING>}. Such breakpoint won't fire until a shared
4005 library that has the symbol or line referred by breakpoint is loaded.
4006 See below for details. A breakpoint with several locations will
4007 have @samp{<MULTIPLE>} in this field---see below for details.
4009 Where the breakpoint is in the source for your program, as a file and
4010 line number. For a pending breakpoint, the original string passed to
4011 the breakpoint command will be listed as it cannot be resolved until
4012 the appropriate shared library is loaded in the future.
4016 If a breakpoint is conditional, there are two evaluation modes: ``host'' and
4017 ``target''. If mode is ``host'', breakpoint condition evaluation is done by
4018 @value{GDBN} on the host's side. If it is ``target'', then the condition
4019 is evaluated by the target. The @code{info break} command shows
4020 the condition on the line following the affected breakpoint, together with
4021 its condition evaluation mode in between parentheses.
4023 Breakpoint commands, if any, are listed after that. A pending breakpoint is
4024 allowed to have a condition specified for it. The condition is not parsed for
4025 validity until a shared library is loaded that allows the pending
4026 breakpoint to resolve to a valid location.
4029 @code{info break} with a breakpoint
4030 number @var{n} as argument lists only that breakpoint. The
4031 convenience variable @code{$_} and the default examining-address for
4032 the @code{x} command are set to the address of the last breakpoint
4033 listed (@pxref{Memory, ,Examining Memory}).
4036 @code{info break} displays a count of the number of times the breakpoint
4037 has been hit. This is especially useful in conjunction with the
4038 @code{ignore} command. You can ignore a large number of breakpoint
4039 hits, look at the breakpoint info to see how many times the breakpoint
4040 was hit, and then run again, ignoring one less than that number. This
4041 will get you quickly to the last hit of that breakpoint.
4044 For a breakpoints with an enable count (xref) greater than 1,
4045 @code{info break} also displays that count.
4049 @value{GDBN} allows you to set any number of breakpoints at the same place in
4050 your program. There is nothing silly or meaningless about this. When
4051 the breakpoints are conditional, this is even useful
4052 (@pxref{Conditions, ,Break Conditions}).
4054 @cindex multiple locations, breakpoints
4055 @cindex breakpoints, multiple locations
4056 It is possible that a breakpoint corresponds to several locations
4057 in your program. Examples of this situation are:
4061 Multiple functions in the program may have the same name.
4064 For a C@t{++} constructor, the @value{NGCC} compiler generates several
4065 instances of the function body, used in different cases.
4068 For a C@t{++} template function, a given line in the function can
4069 correspond to any number of instantiations.
4072 For an inlined function, a given source line can correspond to
4073 several places where that function is inlined.
4076 In all those cases, @value{GDBN} will insert a breakpoint at all
4077 the relevant locations.
4079 A breakpoint with multiple locations is displayed in the breakpoint
4080 table using several rows---one header row, followed by one row for
4081 each breakpoint location. The header row has @samp{<MULTIPLE>} in the
4082 address column. The rows for individual locations contain the actual
4083 addresses for locations, and show the functions to which those
4084 locations belong. The number column for a location is of the form
4085 @var{breakpoint-number}.@var{location-number}.
4090 Num Type Disp Enb Address What
4091 1 breakpoint keep y <MULTIPLE>
4093 breakpoint already hit 1 time
4094 1.1 y 0x080486a2 in void foo<int>() at t.cc:8
4095 1.2 y 0x080486ca in void foo<double>() at t.cc:8
4098 You cannot delete the individual locations from a breakpoint. However,
4099 each location can be individually enabled or disabled by passing
4100 @var{breakpoint-number}.@var{location-number} as argument to the
4101 @code{enable} and @code{disable} commands. It's also possible to
4102 @code{enable} and @code{disable} a range of @var{location-number}
4103 locations using a @var{breakpoint-number} and two @var{location-number}s,
4104 in increasing order, separated by a hyphen, like
4105 @kbd{@var{breakpoint-number}.@var{location-number1}-@var{location-number2}},
4106 in which case @value{GDBN} acts on all the locations in the range (inclusive).
4107 Disabling or enabling the parent breakpoint (@pxref{Disabling}) affects
4108 all of the locations that belong to that breakpoint.
4110 @cindex pending breakpoints
4111 It's quite common to have a breakpoint inside a shared library.
4112 Shared libraries can be loaded and unloaded explicitly,
4113 and possibly repeatedly, as the program is executed. To support
4114 this use case, @value{GDBN} updates breakpoint locations whenever
4115 any shared library is loaded or unloaded. Typically, you would
4116 set a breakpoint in a shared library at the beginning of your
4117 debugging session, when the library is not loaded, and when the
4118 symbols from the library are not available. When you try to set
4119 breakpoint, @value{GDBN} will ask you if you want to set
4120 a so called @dfn{pending breakpoint}---breakpoint whose address
4121 is not yet resolved.
4123 After the program is run, whenever a new shared library is loaded,
4124 @value{GDBN} reevaluates all the breakpoints. When a newly loaded
4125 shared library contains the symbol or line referred to by some
4126 pending breakpoint, that breakpoint is resolved and becomes an
4127 ordinary breakpoint. When a library is unloaded, all breakpoints
4128 that refer to its symbols or source lines become pending again.
4130 This logic works for breakpoints with multiple locations, too. For
4131 example, if you have a breakpoint in a C@t{++} template function, and
4132 a newly loaded shared library has an instantiation of that template,
4133 a new location is added to the list of locations for the breakpoint.
4135 Except for having unresolved address, pending breakpoints do not
4136 differ from regular breakpoints. You can set conditions or commands,
4137 enable and disable them and perform other breakpoint operations.
4139 @value{GDBN} provides some additional commands for controlling what
4140 happens when the @samp{break} command cannot resolve breakpoint
4141 address specification to an address:
4143 @kindex set breakpoint pending
4144 @kindex show breakpoint pending
4146 @item set breakpoint pending auto
4147 This is the default behavior. When @value{GDBN} cannot find the breakpoint
4148 location, it queries you whether a pending breakpoint should be created.
4150 @item set breakpoint pending on
4151 This indicates that an unrecognized breakpoint location should automatically
4152 result in a pending breakpoint being created.
4154 @item set breakpoint pending off
4155 This indicates that pending breakpoints are not to be created. Any
4156 unrecognized breakpoint location results in an error. This setting does
4157 not affect any pending breakpoints previously created.
4159 @item show breakpoint pending
4160 Show the current behavior setting for creating pending breakpoints.
4163 The settings above only affect the @code{break} command and its
4164 variants. Once breakpoint is set, it will be automatically updated
4165 as shared libraries are loaded and unloaded.
4167 @cindex automatic hardware breakpoints
4168 For some targets, @value{GDBN} can automatically decide if hardware or
4169 software breakpoints should be used, depending on whether the
4170 breakpoint address is read-only or read-write. This applies to
4171 breakpoints set with the @code{break} command as well as to internal
4172 breakpoints set by commands like @code{next} and @code{finish}. For
4173 breakpoints set with @code{hbreak}, @value{GDBN} will always use hardware
4176 You can control this automatic behaviour with the following commands:
4178 @kindex set breakpoint auto-hw
4179 @kindex show breakpoint auto-hw
4181 @item set breakpoint auto-hw on
4182 This is the default behavior. When @value{GDBN} sets a breakpoint, it
4183 will try to use the target memory map to decide if software or hardware
4184 breakpoint must be used.
4186 @item set breakpoint auto-hw off
4187 This indicates @value{GDBN} should not automatically select breakpoint
4188 type. If the target provides a memory map, @value{GDBN} will warn when
4189 trying to set software breakpoint at a read-only address.
4192 @value{GDBN} normally implements breakpoints by replacing the program code
4193 at the breakpoint address with a special instruction, which, when
4194 executed, given control to the debugger. By default, the program
4195 code is so modified only when the program is resumed. As soon as
4196 the program stops, @value{GDBN} restores the original instructions. This
4197 behaviour guards against leaving breakpoints inserted in the
4198 target should gdb abrubptly disconnect. However, with slow remote
4199 targets, inserting and removing breakpoint can reduce the performance.
4200 This behavior can be controlled with the following commands::
4202 @kindex set breakpoint always-inserted
4203 @kindex show breakpoint always-inserted
4205 @item set breakpoint always-inserted off
4206 All breakpoints, including newly added by the user, are inserted in
4207 the target only when the target is resumed. All breakpoints are
4208 removed from the target when it stops. This is the default mode.
4210 @item set breakpoint always-inserted on
4211 Causes all breakpoints to be inserted in the target at all times. If
4212 the user adds a new breakpoint, or changes an existing breakpoint, the
4213 breakpoints in the target are updated immediately. A breakpoint is
4214 removed from the target only when breakpoint itself is deleted.
4217 @value{GDBN} handles conditional breakpoints by evaluating these conditions
4218 when a breakpoint breaks. If the condition is true, then the process being
4219 debugged stops, otherwise the process is resumed.
4221 If the target supports evaluating conditions on its end, @value{GDBN} may
4222 download the breakpoint, together with its conditions, to it.
4224 This feature can be controlled via the following commands:
4226 @kindex set breakpoint condition-evaluation
4227 @kindex show breakpoint condition-evaluation
4229 @item set breakpoint condition-evaluation host
4230 This option commands @value{GDBN} to evaluate the breakpoint
4231 conditions on the host's side. Unconditional breakpoints are sent to
4232 the target which in turn receives the triggers and reports them back to GDB
4233 for condition evaluation. This is the standard evaluation mode.
4235 @item set breakpoint condition-evaluation target
4236 This option commands @value{GDBN} to download breakpoint conditions
4237 to the target at the moment of their insertion. The target
4238 is responsible for evaluating the conditional expression and reporting
4239 breakpoint stop events back to @value{GDBN} whenever the condition
4240 is true. Due to limitations of target-side evaluation, some conditions
4241 cannot be evaluated there, e.g., conditions that depend on local data
4242 that is only known to the host. Examples include
4243 conditional expressions involving convenience variables, complex types
4244 that cannot be handled by the agent expression parser and expressions
4245 that are too long to be sent over to the target, specially when the
4246 target is a remote system. In these cases, the conditions will be
4247 evaluated by @value{GDBN}.
4249 @item set breakpoint condition-evaluation auto
4250 This is the default mode. If the target supports evaluating breakpoint
4251 conditions on its end, @value{GDBN} will download breakpoint conditions to
4252 the target (limitations mentioned previously apply). If the target does
4253 not support breakpoint condition evaluation, then @value{GDBN} will fallback
4254 to evaluating all these conditions on the host's side.
4258 @cindex negative breakpoint numbers
4259 @cindex internal @value{GDBN} breakpoints
4260 @value{GDBN} itself sometimes sets breakpoints in your program for
4261 special purposes, such as proper handling of @code{longjmp} (in C
4262 programs). These internal breakpoints are assigned negative numbers,
4263 starting with @code{-1}; @samp{info breakpoints} does not display them.
4264 You can see these breakpoints with the @value{GDBN} maintenance command
4265 @samp{maint info breakpoints} (@pxref{maint info breakpoints}).
4268 @node Set Watchpoints
4269 @subsection Setting Watchpoints
4271 @cindex setting watchpoints
4272 You can use a watchpoint to stop execution whenever the value of an
4273 expression changes, without having to predict a particular place where
4274 this may happen. (This is sometimes called a @dfn{data breakpoint}.)
4275 The expression may be as simple as the value of a single variable, or
4276 as complex as many variables combined by operators. Examples include:
4280 A reference to the value of a single variable.
4283 An address cast to an appropriate data type. For example,
4284 @samp{*(int *)0x12345678} will watch a 4-byte region at the specified
4285 address (assuming an @code{int} occupies 4 bytes).
4288 An arbitrarily complex expression, such as @samp{a*b + c/d}. The
4289 expression can use any operators valid in the program's native
4290 language (@pxref{Languages}).
4293 You can set a watchpoint on an expression even if the expression can
4294 not be evaluated yet. For instance, you can set a watchpoint on
4295 @samp{*global_ptr} before @samp{global_ptr} is initialized.
4296 @value{GDBN} will stop when your program sets @samp{global_ptr} and
4297 the expression produces a valid value. If the expression becomes
4298 valid in some other way than changing a variable (e.g.@: if the memory
4299 pointed to by @samp{*global_ptr} becomes readable as the result of a
4300 @code{malloc} call), @value{GDBN} may not stop until the next time
4301 the expression changes.
4303 @cindex software watchpoints
4304 @cindex hardware watchpoints
4305 Depending on your system, watchpoints may be implemented in software or
4306 hardware. @value{GDBN} does software watchpointing by single-stepping your
4307 program and testing the variable's value each time, which is hundreds of
4308 times slower than normal execution. (But this may still be worth it, to
4309 catch errors where you have no clue what part of your program is the
4312 On some systems, such as most PowerPC or x86-based targets,
4313 @value{GDBN} includes support for hardware watchpoints, which do not
4314 slow down the running of your program.
4318 @item watch @r{[}-l@r{|}-location@r{]} @var{expr} @r{[}thread @var{thread-id}@r{]} @r{[}mask @var{maskvalue}@r{]}
4319 Set a watchpoint for an expression. @value{GDBN} will break when the
4320 expression @var{expr} is written into by the program and its value
4321 changes. The simplest (and the most popular) use of this command is
4322 to watch the value of a single variable:
4325 (@value{GDBP}) watch foo
4328 If the command includes a @code{@r{[}thread @var{thread-id}@r{]}}
4329 argument, @value{GDBN} breaks only when the thread identified by
4330 @var{thread-id} changes the value of @var{expr}. If any other threads
4331 change the value of @var{expr}, @value{GDBN} will not break. Note
4332 that watchpoints restricted to a single thread in this way only work
4333 with Hardware Watchpoints.
4335 Ordinarily a watchpoint respects the scope of variables in @var{expr}
4336 (see below). The @code{-location} argument tells @value{GDBN} to
4337 instead watch the memory referred to by @var{expr}. In this case,
4338 @value{GDBN} will evaluate @var{expr}, take the address of the result,
4339 and watch the memory at that address. The type of the result is used
4340 to determine the size of the watched memory. If the expression's
4341 result does not have an address, then @value{GDBN} will print an
4344 The @code{@r{[}mask @var{maskvalue}@r{]}} argument allows creation
4345 of masked watchpoints, if the current architecture supports this
4346 feature (e.g., PowerPC Embedded architecture, see @ref{PowerPC
4347 Embedded}.) A @dfn{masked watchpoint} specifies a mask in addition
4348 to an address to watch. The mask specifies that some bits of an address
4349 (the bits which are reset in the mask) should be ignored when matching
4350 the address accessed by the inferior against the watchpoint address.
4351 Thus, a masked watchpoint watches many addresses simultaneously---those
4352 addresses whose unmasked bits are identical to the unmasked bits in the
4353 watchpoint address. The @code{mask} argument implies @code{-location}.
4357 (@value{GDBP}) watch foo mask 0xffff00ff
4358 (@value{GDBP}) watch *0xdeadbeef mask 0xffffff00
4362 @item rwatch @r{[}-l@r{|}-location@r{]} @var{expr} @r{[}thread @var{thread-id}@r{]} @r{[}mask @var{maskvalue}@r{]}
4363 Set a watchpoint that will break when the value of @var{expr} is read
4367 @item awatch @r{[}-l@r{|}-location@r{]} @var{expr} @r{[}thread @var{thread-id}@r{]} @r{[}mask @var{maskvalue}@r{]}
4368 Set a watchpoint that will break when @var{expr} is either read from
4369 or written into by the program.
4371 @kindex info watchpoints @r{[}@var{list}@dots{}@r{]}
4372 @item info watchpoints @r{[}@var{list}@dots{}@r{]}
4373 This command prints a list of watchpoints, using the same format as
4374 @code{info break} (@pxref{Set Breaks}).
4377 If you watch for a change in a numerically entered address you need to
4378 dereference it, as the address itself is just a constant number which will
4379 never change. @value{GDBN} refuses to create a watchpoint that watches
4380 a never-changing value:
4383 (@value{GDBP}) watch 0x600850
4384 Cannot watch constant value 0x600850.
4385 (@value{GDBP}) watch *(int *) 0x600850
4386 Watchpoint 1: *(int *) 6293584
4389 @value{GDBN} sets a @dfn{hardware watchpoint} if possible. Hardware
4390 watchpoints execute very quickly, and the debugger reports a change in
4391 value at the exact instruction where the change occurs. If @value{GDBN}
4392 cannot set a hardware watchpoint, it sets a software watchpoint, which
4393 executes more slowly and reports the change in value at the next
4394 @emph{statement}, not the instruction, after the change occurs.
4396 @cindex use only software watchpoints
4397 You can force @value{GDBN} to use only software watchpoints with the
4398 @kbd{set can-use-hw-watchpoints 0} command. With this variable set to
4399 zero, @value{GDBN} will never try to use hardware watchpoints, even if
4400 the underlying system supports them. (Note that hardware-assisted
4401 watchpoints that were set @emph{before} setting
4402 @code{can-use-hw-watchpoints} to zero will still use the hardware
4403 mechanism of watching expression values.)
4406 @item set can-use-hw-watchpoints
4407 @kindex set can-use-hw-watchpoints
4408 Set whether or not to use hardware watchpoints.
4410 @item show can-use-hw-watchpoints
4411 @kindex show can-use-hw-watchpoints
4412 Show the current mode of using hardware watchpoints.
4415 For remote targets, you can restrict the number of hardware
4416 watchpoints @value{GDBN} will use, see @ref{set remote
4417 hardware-breakpoint-limit}.
4419 When you issue the @code{watch} command, @value{GDBN} reports
4422 Hardware watchpoint @var{num}: @var{expr}
4426 if it was able to set a hardware watchpoint.
4428 Currently, the @code{awatch} and @code{rwatch} commands can only set
4429 hardware watchpoints, because accesses to data that don't change the
4430 value of the watched expression cannot be detected without examining
4431 every instruction as it is being executed, and @value{GDBN} does not do
4432 that currently. If @value{GDBN} finds that it is unable to set a
4433 hardware breakpoint with the @code{awatch} or @code{rwatch} command, it
4434 will print a message like this:
4437 Expression cannot be implemented with read/access watchpoint.
4440 Sometimes, @value{GDBN} cannot set a hardware watchpoint because the
4441 data type of the watched expression is wider than what a hardware
4442 watchpoint on the target machine can handle. For example, some systems
4443 can only watch regions that are up to 4 bytes wide; on such systems you
4444 cannot set hardware watchpoints for an expression that yields a
4445 double-precision floating-point number (which is typically 8 bytes
4446 wide). As a work-around, it might be possible to break the large region
4447 into a series of smaller ones and watch them with separate watchpoints.
4449 If you set too many hardware watchpoints, @value{GDBN} might be unable
4450 to insert all of them when you resume the execution of your program.
4451 Since the precise number of active watchpoints is unknown until such
4452 time as the program is about to be resumed, @value{GDBN} might not be
4453 able to warn you about this when you set the watchpoints, and the
4454 warning will be printed only when the program is resumed:
4457 Hardware watchpoint @var{num}: Could not insert watchpoint
4461 If this happens, delete or disable some of the watchpoints.
4463 Watching complex expressions that reference many variables can also
4464 exhaust the resources available for hardware-assisted watchpoints.
4465 That's because @value{GDBN} needs to watch every variable in the
4466 expression with separately allocated resources.
4468 If you call a function interactively using @code{print} or @code{call},
4469 any watchpoints you have set will be inactive until @value{GDBN} reaches another
4470 kind of breakpoint or the call completes.
4472 @value{GDBN} automatically deletes watchpoints that watch local
4473 (automatic) variables, or expressions that involve such variables, when
4474 they go out of scope, that is, when the execution leaves the block in
4475 which these variables were defined. In particular, when the program
4476 being debugged terminates, @emph{all} local variables go out of scope,
4477 and so only watchpoints that watch global variables remain set. If you
4478 rerun the program, you will need to set all such watchpoints again. One
4479 way of doing that would be to set a code breakpoint at the entry to the
4480 @code{main} function and when it breaks, set all the watchpoints.
4482 @cindex watchpoints and threads
4483 @cindex threads and watchpoints
4484 In multi-threaded programs, watchpoints will detect changes to the
4485 watched expression from every thread.
4488 @emph{Warning:} In multi-threaded programs, software watchpoints
4489 have only limited usefulness. If @value{GDBN} creates a software
4490 watchpoint, it can only watch the value of an expression @emph{in a
4491 single thread}. If you are confident that the expression can only
4492 change due to the current thread's activity (and if you are also
4493 confident that no other thread can become current), then you can use
4494 software watchpoints as usual. However, @value{GDBN} may not notice
4495 when a non-current thread's activity changes the expression. (Hardware
4496 watchpoints, in contrast, watch an expression in all threads.)
4499 @xref{set remote hardware-watchpoint-limit}.
4501 @node Set Catchpoints
4502 @subsection Setting Catchpoints
4503 @cindex catchpoints, setting
4504 @cindex exception handlers
4505 @cindex event handling
4507 You can use @dfn{catchpoints} to cause the debugger to stop for certain
4508 kinds of program events, such as C@t{++} exceptions or the loading of a
4509 shared library. Use the @code{catch} command to set a catchpoint.
4513 @item catch @var{event}
4514 Stop when @var{event} occurs. The @var{event} can be any of the following:
4517 @item throw @r{[}@var{regexp}@r{]}
4518 @itemx rethrow @r{[}@var{regexp}@r{]}
4519 @itemx catch @r{[}@var{regexp}@r{]}
4521 @kindex catch rethrow
4523 @cindex stop on C@t{++} exceptions
4524 The throwing, re-throwing, or catching of a C@t{++} exception.
4526 If @var{regexp} is given, then only exceptions whose type matches the
4527 regular expression will be caught.
4529 @vindex $_exception@r{, convenience variable}
4530 The convenience variable @code{$_exception} is available at an
4531 exception-related catchpoint, on some systems. This holds the
4532 exception being thrown.
4534 There are currently some limitations to C@t{++} exception handling in
4539 The support for these commands is system-dependent. Currently, only
4540 systems using the @samp{gnu-v3} C@t{++} ABI (@pxref{ABI}) are
4544 The regular expression feature and the @code{$_exception} convenience
4545 variable rely on the presence of some SDT probes in @code{libstdc++}.
4546 If these probes are not present, then these features cannot be used.
4547 These probes were first available in the GCC 4.8 release, but whether
4548 or not they are available in your GCC also depends on how it was
4552 The @code{$_exception} convenience variable is only valid at the
4553 instruction at which an exception-related catchpoint is set.
4556 When an exception-related catchpoint is hit, @value{GDBN} stops at a
4557 location in the system library which implements runtime exception
4558 support for C@t{++}, usually @code{libstdc++}. You can use @code{up}
4559 (@pxref{Selection}) to get to your code.
4562 If you call a function interactively, @value{GDBN} normally returns
4563 control to you when the function has finished executing. If the call
4564 raises an exception, however, the call may bypass the mechanism that
4565 returns control to you and cause your program either to abort or to
4566 simply continue running until it hits a breakpoint, catches a signal
4567 that @value{GDBN} is listening for, or exits. This is the case even if
4568 you set a catchpoint for the exception; catchpoints on exceptions are
4569 disabled within interactive calls. @xref{Calling}, for information on
4570 controlling this with @code{set unwind-on-terminating-exception}.
4573 You cannot raise an exception interactively.
4576 You cannot install an exception handler interactively.
4579 @item exception @r{[}@var{name}@r{]}
4580 @kindex catch exception
4581 @cindex Ada exception catching
4582 @cindex catch Ada exceptions
4583 An Ada exception being raised. If an exception name is specified
4584 at the end of the command (eg @code{catch exception Program_Error}),
4585 the debugger will stop only when this specific exception is raised.
4586 Otherwise, the debugger stops execution when any Ada exception is raised.
4588 When inserting an exception catchpoint on a user-defined exception whose
4589 name is identical to one of the exceptions defined by the language, the
4590 fully qualified name must be used as the exception name. Otherwise,
4591 @value{GDBN} will assume that it should stop on the pre-defined exception
4592 rather than the user-defined one. For instance, assuming an exception
4593 called @code{Constraint_Error} is defined in package @code{Pck}, then
4594 the command to use to catch such exceptions is @kbd{catch exception
4595 Pck.Constraint_Error}.
4597 @item exception unhandled
4598 @kindex catch exception unhandled
4599 An exception that was raised but is not handled by the program.
4601 @item handlers @r{[}@var{name}@r{]}
4602 @kindex catch handlers
4603 @cindex Ada exception handlers catching
4604 @cindex catch Ada exceptions when handled
4605 An Ada exception being handled. If an exception name is
4606 specified at the end of the command
4607 (eg @kbd{catch handlers Program_Error}), the debugger will stop
4608 only when this specific exception is handled.
4609 Otherwise, the debugger stops execution when any Ada exception is handled.
4611 When inserting a handlers catchpoint on a user-defined
4612 exception whose name is identical to one of the exceptions
4613 defined by the language, the fully qualified name must be used
4614 as the exception name. Otherwise, @value{GDBN} will assume that it
4615 should stop on the pre-defined exception rather than the
4616 user-defined one. For instance, assuming an exception called
4617 @code{Constraint_Error} is defined in package @code{Pck}, then the
4618 command to use to catch such exceptions handling is
4619 @kbd{catch handlers Pck.Constraint_Error}.
4622 @kindex catch assert
4623 A failed Ada assertion.
4627 @cindex break on fork/exec
4628 A call to @code{exec}.
4630 @anchor{catch syscall}
4632 @itemx syscall @r{[}@var{name} @r{|} @var{number} @r{|} @r{group:}@var{groupname} @r{|} @r{g:}@var{groupname}@r{]} @dots{}
4633 @kindex catch syscall
4634 @cindex break on a system call.
4635 A call to or return from a system call, a.k.a.@: @dfn{syscall}. A
4636 syscall is a mechanism for application programs to request a service
4637 from the operating system (OS) or one of the OS system services.
4638 @value{GDBN} can catch some or all of the syscalls issued by the
4639 debuggee, and show the related information for each syscall. If no
4640 argument is specified, calls to and returns from all system calls
4643 @var{name} can be any system call name that is valid for the
4644 underlying OS. Just what syscalls are valid depends on the OS. On
4645 GNU and Unix systems, you can find the full list of valid syscall
4646 names on @file{/usr/include/asm/unistd.h}.
4648 @c For MS-Windows, the syscall names and the corresponding numbers
4649 @c can be found, e.g., on this URL:
4650 @c http://www.metasploit.com/users/opcode/syscalls.html
4651 @c but we don't support Windows syscalls yet.
4653 Normally, @value{GDBN} knows in advance which syscalls are valid for
4654 each OS, so you can use the @value{GDBN} command-line completion
4655 facilities (@pxref{Completion,, command completion}) to list the
4658 You may also specify the system call numerically. A syscall's
4659 number is the value passed to the OS's syscall dispatcher to
4660 identify the requested service. When you specify the syscall by its
4661 name, @value{GDBN} uses its database of syscalls to convert the name
4662 into the corresponding numeric code, but using the number directly
4663 may be useful if @value{GDBN}'s database does not have the complete
4664 list of syscalls on your system (e.g., because @value{GDBN} lags
4665 behind the OS upgrades).
4667 You may specify a group of related syscalls to be caught at once using
4668 the @code{group:} syntax (@code{g:} is a shorter equivalent). For
4669 instance, on some platforms @value{GDBN} allows you to catch all
4670 network related syscalls, by passing the argument @code{group:network}
4671 to @code{catch syscall}. Note that not all syscall groups are
4672 available in every system. You can use the command completion
4673 facilities (@pxref{Completion,, command completion}) to list the
4674 syscall groups available on your environment.
4676 The example below illustrates how this command works if you don't provide
4680 (@value{GDBP}) catch syscall
4681 Catchpoint 1 (syscall)
4683 Starting program: /tmp/catch-syscall
4685 Catchpoint 1 (call to syscall 'close'), \
4686 0xffffe424 in __kernel_vsyscall ()
4690 Catchpoint 1 (returned from syscall 'close'), \
4691 0xffffe424 in __kernel_vsyscall ()
4695 Here is an example of catching a system call by name:
4698 (@value{GDBP}) catch syscall chroot
4699 Catchpoint 1 (syscall 'chroot' [61])
4701 Starting program: /tmp/catch-syscall
4703 Catchpoint 1 (call to syscall 'chroot'), \
4704 0xffffe424 in __kernel_vsyscall ()
4708 Catchpoint 1 (returned from syscall 'chroot'), \
4709 0xffffe424 in __kernel_vsyscall ()
4713 An example of specifying a system call numerically. In the case
4714 below, the syscall number has a corresponding entry in the XML
4715 file, so @value{GDBN} finds its name and prints it:
4718 (@value{GDBP}) catch syscall 252
4719 Catchpoint 1 (syscall(s) 'exit_group')
4721 Starting program: /tmp/catch-syscall
4723 Catchpoint 1 (call to syscall 'exit_group'), \
4724 0xffffe424 in __kernel_vsyscall ()
4728 Program exited normally.
4732 Here is an example of catching a syscall group:
4735 (@value{GDBP}) catch syscall group:process
4736 Catchpoint 1 (syscalls 'exit' [1] 'fork' [2] 'waitpid' [7]
4737 'execve' [11] 'wait4' [114] 'clone' [120] 'vfork' [190]
4738 'exit_group' [252] 'waitid' [284] 'unshare' [310])
4740 Starting program: /tmp/catch-syscall
4742 Catchpoint 1 (call to syscall fork), 0x00007ffff7df4e27 in open64 ()
4743 from /lib64/ld-linux-x86-64.so.2
4749 However, there can be situations when there is no corresponding name
4750 in XML file for that syscall number. In this case, @value{GDBN} prints
4751 a warning message saying that it was not able to find the syscall name,
4752 but the catchpoint will be set anyway. See the example below:
4755 (@value{GDBP}) catch syscall 764
4756 warning: The number '764' does not represent a known syscall.
4757 Catchpoint 2 (syscall 764)
4761 If you configure @value{GDBN} using the @samp{--without-expat} option,
4762 it will not be able to display syscall names. Also, if your
4763 architecture does not have an XML file describing its system calls,
4764 you will not be able to see the syscall names. It is important to
4765 notice that these two features are used for accessing the syscall
4766 name database. In either case, you will see a warning like this:
4769 (@value{GDBP}) catch syscall
4770 warning: Could not open "syscalls/i386-linux.xml"
4771 warning: Could not load the syscall XML file 'syscalls/i386-linux.xml'.
4772 GDB will not be able to display syscall names.
4773 Catchpoint 1 (syscall)
4777 Of course, the file name will change depending on your architecture and system.
4779 Still using the example above, you can also try to catch a syscall by its
4780 number. In this case, you would see something like:
4783 (@value{GDBP}) catch syscall 252
4784 Catchpoint 1 (syscall(s) 252)
4787 Again, in this case @value{GDBN} would not be able to display syscall's names.
4791 A call to @code{fork}.
4795 A call to @code{vfork}.
4797 @item load @r{[}@var{regexp}@r{]}
4798 @itemx unload @r{[}@var{regexp}@r{]}
4800 @kindex catch unload
4801 The loading or unloading of a shared library. If @var{regexp} is
4802 given, then the catchpoint will stop only if the regular expression
4803 matches one of the affected libraries.
4805 @item signal @r{[}@var{signal}@dots{} @r{|} @samp{all}@r{]}
4806 @kindex catch signal
4807 The delivery of a signal.
4809 With no arguments, this catchpoint will catch any signal that is not
4810 used internally by @value{GDBN}, specifically, all signals except
4811 @samp{SIGTRAP} and @samp{SIGINT}.
4813 With the argument @samp{all}, all signals, including those used by
4814 @value{GDBN}, will be caught. This argument cannot be used with other
4817 Otherwise, the arguments are a list of signal names as given to
4818 @code{handle} (@pxref{Signals}). Only signals specified in this list
4821 One reason that @code{catch signal} can be more useful than
4822 @code{handle} is that you can attach commands and conditions to the
4825 When a signal is caught by a catchpoint, the signal's @code{stop} and
4826 @code{print} settings, as specified by @code{handle}, are ignored.
4827 However, whether the signal is still delivered to the inferior depends
4828 on the @code{pass} setting; this can be changed in the catchpoint's
4833 @item tcatch @var{event}
4835 Set a catchpoint that is enabled only for one stop. The catchpoint is
4836 automatically deleted after the first time the event is caught.
4840 Use the @code{info break} command to list the current catchpoints.
4844 @subsection Deleting Breakpoints
4846 @cindex clearing breakpoints, watchpoints, catchpoints
4847 @cindex deleting breakpoints, watchpoints, catchpoints
4848 It is often necessary to eliminate a breakpoint, watchpoint, or
4849 catchpoint once it has done its job and you no longer want your program
4850 to stop there. This is called @dfn{deleting} the breakpoint. A
4851 breakpoint that has been deleted no longer exists; it is forgotten.
4853 With the @code{clear} command you can delete breakpoints according to
4854 where they are in your program. With the @code{delete} command you can
4855 delete individual breakpoints, watchpoints, or catchpoints by specifying
4856 their breakpoint numbers.
4858 It is not necessary to delete a breakpoint to proceed past it. @value{GDBN}
4859 automatically ignores breakpoints on the first instruction to be executed
4860 when you continue execution without changing the execution address.
4865 Delete any breakpoints at the next instruction to be executed in the
4866 selected stack frame (@pxref{Selection, ,Selecting a Frame}). When
4867 the innermost frame is selected, this is a good way to delete a
4868 breakpoint where your program just stopped.
4870 @item clear @var{location}
4871 Delete any breakpoints set at the specified @var{location}.
4872 @xref{Specify Location}, for the various forms of @var{location}; the
4873 most useful ones are listed below:
4876 @item clear @var{function}
4877 @itemx clear @var{filename}:@var{function}
4878 Delete any breakpoints set at entry to the named @var{function}.
4880 @item clear @var{linenum}
4881 @itemx clear @var{filename}:@var{linenum}
4882 Delete any breakpoints set at or within the code of the specified
4883 @var{linenum} of the specified @var{filename}.
4886 @cindex delete breakpoints
4888 @kindex d @r{(@code{delete})}
4889 @item delete @r{[}breakpoints@r{]} @r{[}@var{list}@dots{}@r{]}
4890 Delete the breakpoints, watchpoints, or catchpoints of the breakpoint
4891 list specified as argument. If no argument is specified, delete all
4892 breakpoints (@value{GDBN} asks confirmation, unless you have @code{set
4893 confirm off}). You can abbreviate this command as @code{d}.
4897 @subsection Disabling Breakpoints
4899 @cindex enable/disable a breakpoint
4900 Rather than deleting a breakpoint, watchpoint, or catchpoint, you might
4901 prefer to @dfn{disable} it. This makes the breakpoint inoperative as if
4902 it had been deleted, but remembers the information on the breakpoint so
4903 that you can @dfn{enable} it again later.
4905 You disable and enable breakpoints, watchpoints, and catchpoints with
4906 the @code{enable} and @code{disable} commands, optionally specifying
4907 one or more breakpoint numbers as arguments. Use @code{info break} to
4908 print a list of all breakpoints, watchpoints, and catchpoints if you
4909 do not know which numbers to use.
4911 Disabling and enabling a breakpoint that has multiple locations
4912 affects all of its locations.
4914 A breakpoint, watchpoint, or catchpoint can have any of several
4915 different states of enablement:
4919 Enabled. The breakpoint stops your program. A breakpoint set
4920 with the @code{break} command starts out in this state.
4922 Disabled. The breakpoint has no effect on your program.
4924 Enabled once. The breakpoint stops your program, but then becomes
4927 Enabled for a count. The breakpoint stops your program for the next
4928 N times, then becomes disabled.
4930 Enabled for deletion. The breakpoint stops your program, but
4931 immediately after it does so it is deleted permanently. A breakpoint
4932 set with the @code{tbreak} command starts out in this state.
4935 You can use the following commands to enable or disable breakpoints,
4936 watchpoints, and catchpoints:
4940 @kindex dis @r{(@code{disable})}
4941 @item disable @r{[}breakpoints@r{]} @r{[}@var{list}@dots{}@r{]}
4942 Disable the specified breakpoints---or all breakpoints, if none are
4943 listed. A disabled breakpoint has no effect but is not forgotten. All
4944 options such as ignore-counts, conditions and commands are remembered in
4945 case the breakpoint is enabled again later. You may abbreviate
4946 @code{disable} as @code{dis}.
4949 @item enable @r{[}breakpoints@r{]} @r{[}@var{list}@dots{}@r{]}
4950 Enable the specified breakpoints (or all defined breakpoints). They
4951 become effective once again in stopping your program.
4953 @item enable @r{[}breakpoints@r{]} once @var{list}@dots{}
4954 Enable the specified breakpoints temporarily. @value{GDBN} disables any
4955 of these breakpoints immediately after stopping your program.
4957 @item enable @r{[}breakpoints@r{]} count @var{count} @var{list}@dots{}
4958 Enable the specified breakpoints temporarily. @value{GDBN} records
4959 @var{count} with each of the specified breakpoints, and decrements a
4960 breakpoint's count when it is hit. When any count reaches 0,
4961 @value{GDBN} disables that breakpoint. If a breakpoint has an ignore
4962 count (@pxref{Conditions, ,Break Conditions}), that will be
4963 decremented to 0 before @var{count} is affected.
4965 @item enable @r{[}breakpoints@r{]} delete @var{list}@dots{}
4966 Enable the specified breakpoints to work once, then die. @value{GDBN}
4967 deletes any of these breakpoints as soon as your program stops there.
4968 Breakpoints set by the @code{tbreak} command start out in this state.
4971 @c FIXME: I think the following ``Except for [...] @code{tbreak}'' is
4972 @c confusing: tbreak is also initially enabled.
4973 Except for a breakpoint set with @code{tbreak} (@pxref{Set Breaks,
4974 ,Setting Breakpoints}), breakpoints that you set are initially enabled;
4975 subsequently, they become disabled or enabled only when you use one of
4976 the commands above. (The command @code{until} can set and delete a
4977 breakpoint of its own, but it does not change the state of your other
4978 breakpoints; see @ref{Continuing and Stepping, ,Continuing and
4982 @subsection Break Conditions
4983 @cindex conditional breakpoints
4984 @cindex breakpoint conditions
4986 @c FIXME what is scope of break condition expr? Context where wanted?
4987 @c in particular for a watchpoint?
4988 The simplest sort of breakpoint breaks every time your program reaches a
4989 specified place. You can also specify a @dfn{condition} for a
4990 breakpoint. A condition is just a Boolean expression in your
4991 programming language (@pxref{Expressions, ,Expressions}). A breakpoint with
4992 a condition evaluates the expression each time your program reaches it,
4993 and your program stops only if the condition is @emph{true}.
4995 This is the converse of using assertions for program validation; in that
4996 situation, you want to stop when the assertion is violated---that is,
4997 when the condition is false. In C, if you want to test an assertion expressed
4998 by the condition @var{assert}, you should set the condition
4999 @samp{! @var{assert}} on the appropriate breakpoint.
5001 Conditions are also accepted for watchpoints; you may not need them,
5002 since a watchpoint is inspecting the value of an expression anyhow---but
5003 it might be simpler, say, to just set a watchpoint on a variable name,
5004 and specify a condition that tests whether the new value is an interesting
5007 Break conditions can have side effects, and may even call functions in
5008 your program. This can be useful, for example, to activate functions
5009 that log program progress, or to use your own print functions to
5010 format special data structures. The effects are completely predictable
5011 unless there is another enabled breakpoint at the same address. (In
5012 that case, @value{GDBN} might see the other breakpoint first and stop your
5013 program without checking the condition of this one.) Note that
5014 breakpoint commands are usually more convenient and flexible than break
5016 purpose of performing side effects when a breakpoint is reached
5017 (@pxref{Break Commands, ,Breakpoint Command Lists}).
5019 Breakpoint conditions can also be evaluated on the target's side if
5020 the target supports it. Instead of evaluating the conditions locally,
5021 @value{GDBN} encodes the expression into an agent expression
5022 (@pxref{Agent Expressions}) suitable for execution on the target,
5023 independently of @value{GDBN}. Global variables become raw memory
5024 locations, locals become stack accesses, and so forth.
5026 In this case, @value{GDBN} will only be notified of a breakpoint trigger
5027 when its condition evaluates to true. This mechanism may provide faster
5028 response times depending on the performance characteristics of the target
5029 since it does not need to keep @value{GDBN} informed about
5030 every breakpoint trigger, even those with false conditions.
5032 Break conditions can be specified when a breakpoint is set, by using
5033 @samp{if} in the arguments to the @code{break} command. @xref{Set
5034 Breaks, ,Setting Breakpoints}. They can also be changed at any time
5035 with the @code{condition} command.
5037 You can also use the @code{if} keyword with the @code{watch} command.
5038 The @code{catch} command does not recognize the @code{if} keyword;
5039 @code{condition} is the only way to impose a further condition on a
5044 @item condition @var{bnum} @var{expression}
5045 Specify @var{expression} as the break condition for breakpoint,
5046 watchpoint, or catchpoint number @var{bnum}. After you set a condition,
5047 breakpoint @var{bnum} stops your program only if the value of
5048 @var{expression} is true (nonzero, in C). When you use
5049 @code{condition}, @value{GDBN} checks @var{expression} immediately for
5050 syntactic correctness, and to determine whether symbols in it have
5051 referents in the context of your breakpoint. If @var{expression} uses
5052 symbols not referenced in the context of the breakpoint, @value{GDBN}
5053 prints an error message:
5056 No symbol "foo" in current context.
5061 not actually evaluate @var{expression} at the time the @code{condition}
5062 command (or a command that sets a breakpoint with a condition, like
5063 @code{break if @dots{}}) is given, however. @xref{Expressions, ,Expressions}.
5065 @item condition @var{bnum}
5066 Remove the condition from breakpoint number @var{bnum}. It becomes
5067 an ordinary unconditional breakpoint.
5070 @cindex ignore count (of breakpoint)
5071 A special case of a breakpoint condition is to stop only when the
5072 breakpoint has been reached a certain number of times. This is so
5073 useful that there is a special way to do it, using the @dfn{ignore
5074 count} of the breakpoint. Every breakpoint has an ignore count, which
5075 is an integer. Most of the time, the ignore count is zero, and
5076 therefore has no effect. But if your program reaches a breakpoint whose
5077 ignore count is positive, then instead of stopping, it just decrements
5078 the ignore count by one and continues. As a result, if the ignore count
5079 value is @var{n}, the breakpoint does not stop the next @var{n} times
5080 your program reaches it.
5084 @item ignore @var{bnum} @var{count}
5085 Set the ignore count of breakpoint number @var{bnum} to @var{count}.
5086 The next @var{count} times the breakpoint is reached, your program's
5087 execution does not stop; other than to decrement the ignore count, @value{GDBN}
5090 To make the breakpoint stop the next time it is reached, specify
5093 When you use @code{continue} to resume execution of your program from a
5094 breakpoint, you can specify an ignore count directly as an argument to
5095 @code{continue}, rather than using @code{ignore}. @xref{Continuing and
5096 Stepping,,Continuing and Stepping}.
5098 If a breakpoint has a positive ignore count and a condition, the
5099 condition is not checked. Once the ignore count reaches zero,
5100 @value{GDBN} resumes checking the condition.
5102 You could achieve the effect of the ignore count with a condition such
5103 as @w{@samp{$foo-- <= 0}} using a debugger convenience variable that
5104 is decremented each time. @xref{Convenience Vars, ,Convenience
5108 Ignore counts apply to breakpoints, watchpoints, and catchpoints.
5111 @node Break Commands
5112 @subsection Breakpoint Command Lists
5114 @cindex breakpoint commands
5115 You can give any breakpoint (or watchpoint or catchpoint) a series of
5116 commands to execute when your program stops due to that breakpoint. For
5117 example, you might want to print the values of certain expressions, or
5118 enable other breakpoints.
5122 @kindex end@r{ (breakpoint commands)}
5123 @item commands @r{[}@var{list}@dots{}@r{]}
5124 @itemx @dots{} @var{command-list} @dots{}
5126 Specify a list of commands for the given breakpoints. The commands
5127 themselves appear on the following lines. Type a line containing just
5128 @code{end} to terminate the commands.
5130 To remove all commands from a breakpoint, type @code{commands} and
5131 follow it immediately with @code{end}; that is, give no commands.
5133 With no argument, @code{commands} refers to the last breakpoint,
5134 watchpoint, or catchpoint set (not to the breakpoint most recently
5135 encountered). If the most recent breakpoints were set with a single
5136 command, then the @code{commands} will apply to all the breakpoints
5137 set by that command. This applies to breakpoints set by
5138 @code{rbreak}, and also applies when a single @code{break} command
5139 creates multiple breakpoints (@pxref{Ambiguous Expressions,,Ambiguous
5143 Pressing @key{RET} as a means of repeating the last @value{GDBN} command is
5144 disabled within a @var{command-list}.
5146 You can use breakpoint commands to start your program up again. Simply
5147 use the @code{continue} command, or @code{step}, or any other command
5148 that resumes execution.
5150 Any other commands in the command list, after a command that resumes
5151 execution, are ignored. This is because any time you resume execution
5152 (even with a simple @code{next} or @code{step}), you may encounter
5153 another breakpoint---which could have its own command list, leading to
5154 ambiguities about which list to execute.
5157 If the first command you specify in a command list is @code{silent}, the
5158 usual message about stopping at a breakpoint is not printed. This may
5159 be desirable for breakpoints that are to print a specific message and
5160 then continue. If none of the remaining commands print anything, you
5161 see no sign that the breakpoint was reached. @code{silent} is
5162 meaningful only at the beginning of a breakpoint command list.
5164 The commands @code{echo}, @code{output}, and @code{printf} allow you to
5165 print precisely controlled output, and are often useful in silent
5166 breakpoints. @xref{Output, ,Commands for Controlled Output}.
5168 For example, here is how you could use breakpoint commands to print the
5169 value of @code{x} at entry to @code{foo} whenever @code{x} is positive.
5175 printf "x is %d\n",x
5180 One application for breakpoint commands is to compensate for one bug so
5181 you can test for another. Put a breakpoint just after the erroneous line
5182 of code, give it a condition to detect the case in which something
5183 erroneous has been done, and give it commands to assign correct values
5184 to any variables that need them. End with the @code{continue} command
5185 so that your program does not stop, and start with the @code{silent}
5186 command so that no output is produced. Here is an example:
5197 @node Dynamic Printf
5198 @subsection Dynamic Printf
5200 @cindex dynamic printf
5202 The dynamic printf command @code{dprintf} combines a breakpoint with
5203 formatted printing of your program's data to give you the effect of
5204 inserting @code{printf} calls into your program on-the-fly, without
5205 having to recompile it.
5207 In its most basic form, the output goes to the GDB console. However,
5208 you can set the variable @code{dprintf-style} for alternate handling.
5209 For instance, you can ask to format the output by calling your
5210 program's @code{printf} function. This has the advantage that the
5211 characters go to the program's output device, so they can recorded in
5212 redirects to files and so forth.
5214 If you are doing remote debugging with a stub or agent, you can also
5215 ask to have the printf handled by the remote agent. In addition to
5216 ensuring that the output goes to the remote program's device along
5217 with any other output the program might produce, you can also ask that
5218 the dprintf remain active even after disconnecting from the remote
5219 target. Using the stub/agent is also more efficient, as it can do
5220 everything without needing to communicate with @value{GDBN}.
5224 @item dprintf @var{location},@var{template},@var{expression}[,@var{expression}@dots{}]
5225 Whenever execution reaches @var{location}, print the values of one or
5226 more @var{expressions} under the control of the string @var{template}.
5227 To print several values, separate them with commas.
5229 @item set dprintf-style @var{style}
5230 Set the dprintf output to be handled in one of several different
5231 styles enumerated below. A change of style affects all existing
5232 dynamic printfs immediately. (If you need individual control over the
5233 print commands, simply define normal breakpoints with
5234 explicitly-supplied command lists.)
5238 @kindex dprintf-style gdb
5239 Handle the output using the @value{GDBN} @code{printf} command.
5242 @kindex dprintf-style call
5243 Handle the output by calling a function in your program (normally
5247 @kindex dprintf-style agent
5248 Have the remote debugging agent (such as @code{gdbserver}) handle
5249 the output itself. This style is only available for agents that
5250 support running commands on the target.
5253 @item set dprintf-function @var{function}
5254 Set the function to call if the dprintf style is @code{call}. By
5255 default its value is @code{printf}. You may set it to any expression.
5256 that @value{GDBN} can evaluate to a function, as per the @code{call}
5259 @item set dprintf-channel @var{channel}
5260 Set a ``channel'' for dprintf. If set to a non-empty value,
5261 @value{GDBN} will evaluate it as an expression and pass the result as
5262 a first argument to the @code{dprintf-function}, in the manner of
5263 @code{fprintf} and similar functions. Otherwise, the dprintf format
5264 string will be the first argument, in the manner of @code{printf}.
5266 As an example, if you wanted @code{dprintf} output to go to a logfile
5267 that is a standard I/O stream assigned to the variable @code{mylog},
5268 you could do the following:
5271 (gdb) set dprintf-style call
5272 (gdb) set dprintf-function fprintf
5273 (gdb) set dprintf-channel mylog
5274 (gdb) dprintf 25,"at line 25, glob=%d\n",glob
5275 Dprintf 1 at 0x123456: file main.c, line 25.
5277 1 dprintf keep y 0x00123456 in main at main.c:25
5278 call (void) fprintf (mylog,"at line 25, glob=%d\n",glob)
5283 Note that the @code{info break} displays the dynamic printf commands
5284 as normal breakpoint commands; you can thus easily see the effect of
5285 the variable settings.
5287 @item set disconnected-dprintf on
5288 @itemx set disconnected-dprintf off
5289 @kindex set disconnected-dprintf
5290 Choose whether @code{dprintf} commands should continue to run if
5291 @value{GDBN} has disconnected from the target. This only applies
5292 if the @code{dprintf-style} is @code{agent}.
5294 @item show disconnected-dprintf off
5295 @kindex show disconnected-dprintf
5296 Show the current choice for disconnected @code{dprintf}.
5300 @value{GDBN} does not check the validity of function and channel,
5301 relying on you to supply values that are meaningful for the contexts
5302 in which they are being used. For instance, the function and channel
5303 may be the values of local variables, but if that is the case, then
5304 all enabled dynamic prints must be at locations within the scope of
5305 those locals. If evaluation fails, @value{GDBN} will report an error.
5307 @node Save Breakpoints
5308 @subsection How to save breakpoints to a file
5310 To save breakpoint definitions to a file use the @w{@code{save
5311 breakpoints}} command.
5314 @kindex save breakpoints
5315 @cindex save breakpoints to a file for future sessions
5316 @item save breakpoints [@var{filename}]
5317 This command saves all current breakpoint definitions together with
5318 their commands and ignore counts, into a file @file{@var{filename}}
5319 suitable for use in a later debugging session. This includes all
5320 types of breakpoints (breakpoints, watchpoints, catchpoints,
5321 tracepoints). To read the saved breakpoint definitions, use the
5322 @code{source} command (@pxref{Command Files}). Note that watchpoints
5323 with expressions involving local variables may fail to be recreated
5324 because it may not be possible to access the context where the
5325 watchpoint is valid anymore. Because the saved breakpoint definitions
5326 are simply a sequence of @value{GDBN} commands that recreate the
5327 breakpoints, you can edit the file in your favorite editing program,
5328 and remove the breakpoint definitions you're not interested in, or
5329 that can no longer be recreated.
5332 @node Static Probe Points
5333 @subsection Static Probe Points
5335 @cindex static probe point, SystemTap
5336 @cindex static probe point, DTrace
5337 @value{GDBN} supports @dfn{SDT} probes in the code. @acronym{SDT} stands
5338 for Statically Defined Tracing, and the probes are designed to have a tiny
5339 runtime code and data footprint, and no dynamic relocations.
5341 Currently, the following types of probes are supported on
5342 ELF-compatible systems:
5346 @item @code{SystemTap} (@uref{http://sourceware.org/systemtap/})
5347 @acronym{SDT} probes@footnote{See
5348 @uref{http://sourceware.org/systemtap/wiki/AddingUserSpaceProbingToApps}
5349 for more information on how to add @code{SystemTap} @acronym{SDT}
5350 probes in your applications.}. @code{SystemTap} probes are usable
5351 from assembly, C and C@t{++} languages@footnote{See
5352 @uref{http://sourceware.org/systemtap/wiki/UserSpaceProbeImplementation}
5353 for a good reference on how the @acronym{SDT} probes are implemented.}.
5355 @item @code{DTrace} (@uref{http://oss.oracle.com/projects/DTrace})
5356 @acronym{USDT} probes. @code{DTrace} probes are usable from C and
5360 @cindex semaphores on static probe points
5361 Some @code{SystemTap} probes have an associated semaphore variable;
5362 for instance, this happens automatically if you defined your probe
5363 using a DTrace-style @file{.d} file. If your probe has a semaphore,
5364 @value{GDBN} will automatically enable it when you specify a
5365 breakpoint using the @samp{-probe-stap} notation. But, if you put a
5366 breakpoint at a probe's location by some other method (e.g.,
5367 @code{break file:line}), then @value{GDBN} will not automatically set
5368 the semaphore. @code{DTrace} probes do not support semaphores.
5370 You can examine the available static static probes using @code{info
5371 probes}, with optional arguments:
5375 @item info probes @r{[}@var{type}@r{]} @r{[}@var{provider} @r{[}@var{name} @r{[}@var{objfile}@r{]}@r{]}@r{]}
5376 If given, @var{type} is either @code{stap} for listing
5377 @code{SystemTap} probes or @code{dtrace} for listing @code{DTrace}
5378 probes. If omitted all probes are listed regardless of their types.
5380 If given, @var{provider} is a regular expression used to match against provider
5381 names when selecting which probes to list. If omitted, probes by all
5382 probes from all providers are listed.
5384 If given, @var{name} is a regular expression to match against probe names
5385 when selecting which probes to list. If omitted, probe names are not
5386 considered when deciding whether to display them.
5388 If given, @var{objfile} is a regular expression used to select which
5389 object files (executable or shared libraries) to examine. If not
5390 given, all object files are considered.
5392 @item info probes all
5393 List the available static probes, from all types.
5396 @cindex enabling and disabling probes
5397 Some probe points can be enabled and/or disabled. The effect of
5398 enabling or disabling a probe depends on the type of probe being
5399 handled. Some @code{DTrace} probes can be enabled or
5400 disabled, but @code{SystemTap} probes cannot be disabled.
5402 You can enable (or disable) one or more probes using the following
5403 commands, with optional arguments:
5406 @kindex enable probes
5407 @item enable probes @r{[}@var{provider} @r{[}@var{name} @r{[}@var{objfile}@r{]}@r{]}@r{]}
5408 If given, @var{provider} is a regular expression used to match against
5409 provider names when selecting which probes to enable. If omitted,
5410 all probes from all providers are enabled.
5412 If given, @var{name} is a regular expression to match against probe
5413 names when selecting which probes to enable. If omitted, probe names
5414 are not considered when deciding whether to enable 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 @kindex disable probes
5421 @item disable probes @r{[}@var{provider} @r{[}@var{name} @r{[}@var{objfile}@r{]}@r{]}@r{]}
5422 See the @code{enable probes} command above for a description of the
5423 optional arguments accepted by this command.
5426 @vindex $_probe_arg@r{, convenience variable}
5427 A probe may specify up to twelve arguments. These are available at the
5428 point at which the probe is defined---that is, when the current PC is
5429 at the probe's location. The arguments are available using the
5430 convenience variables (@pxref{Convenience Vars})
5431 @code{$_probe_arg0}@dots{}@code{$_probe_arg11}. In @code{SystemTap}
5432 probes each probe argument is an integer of the appropriate size;
5433 types are not preserved. In @code{DTrace} probes types are preserved
5434 provided that they are recognized as such by @value{GDBN}; otherwise
5435 the value of the probe argument will be a long integer. The
5436 convenience variable @code{$_probe_argc} holds the number of arguments
5437 at the current probe point.
5439 These variables are always available, but attempts to access them at
5440 any location other than a probe point will cause @value{GDBN} to give
5444 @c @ifclear BARETARGET
5445 @node Error in Breakpoints
5446 @subsection ``Cannot insert breakpoints''
5448 If you request too many active hardware-assisted breakpoints and
5449 watchpoints, you will see this error message:
5451 @c FIXME: the precise wording of this message may change; the relevant
5452 @c source change is not committed yet (Sep 3, 1999).
5454 Stopped; cannot insert breakpoints.
5455 You may have requested too many hardware breakpoints and watchpoints.
5459 This message is printed when you attempt to resume the program, since
5460 only then @value{GDBN} knows exactly how many hardware breakpoints and
5461 watchpoints it needs to insert.
5463 When this message is printed, you need to disable or remove some of the
5464 hardware-assisted breakpoints and watchpoints, and then continue.
5466 @node Breakpoint-related Warnings
5467 @subsection ``Breakpoint address adjusted...''
5468 @cindex breakpoint address adjusted
5470 Some processor architectures place constraints on the addresses at
5471 which breakpoints may be placed. For architectures thus constrained,
5472 @value{GDBN} will attempt to adjust the breakpoint's address to comply
5473 with the constraints dictated by the architecture.
5475 One example of such an architecture is the Fujitsu FR-V. The FR-V is
5476 a VLIW architecture in which a number of RISC-like instructions may be
5477 bundled together for parallel execution. The FR-V architecture
5478 constrains the location of a breakpoint instruction within such a
5479 bundle to the instruction with the lowest address. @value{GDBN}
5480 honors this constraint by adjusting a breakpoint's address to the
5481 first in the bundle.
5483 It is not uncommon for optimized code to have bundles which contain
5484 instructions from different source statements, thus it may happen that
5485 a breakpoint's address will be adjusted from one source statement to
5486 another. Since this adjustment may significantly alter @value{GDBN}'s
5487 breakpoint related behavior from what the user expects, a warning is
5488 printed when the breakpoint is first set and also when the breakpoint
5491 A warning like the one below is printed when setting a breakpoint
5492 that's been subject to address adjustment:
5495 warning: Breakpoint address adjusted from 0x00010414 to 0x00010410.
5498 Such warnings are printed both for user settable and @value{GDBN}'s
5499 internal breakpoints. If you see one of these warnings, you should
5500 verify that a breakpoint set at the adjusted address will have the
5501 desired affect. If not, the breakpoint in question may be removed and
5502 other breakpoints may be set which will have the desired behavior.
5503 E.g., it may be sufficient to place the breakpoint at a later
5504 instruction. A conditional breakpoint may also be useful in some
5505 cases to prevent the breakpoint from triggering too often.
5507 @value{GDBN} will also issue a warning when stopping at one of these
5508 adjusted breakpoints:
5511 warning: Breakpoint 1 address previously adjusted from 0x00010414
5515 When this warning is encountered, it may be too late to take remedial
5516 action except in cases where the breakpoint is hit earlier or more
5517 frequently than expected.
5519 @node Continuing and Stepping
5520 @section Continuing and Stepping
5524 @cindex resuming execution
5525 @dfn{Continuing} means resuming program execution until your program
5526 completes normally. In contrast, @dfn{stepping} means executing just
5527 one more ``step'' of your program, where ``step'' may mean either one
5528 line of source code, or one machine instruction (depending on what
5529 particular command you use). Either when continuing or when stepping,
5530 your program may stop even sooner, due to a breakpoint or a signal. (If
5531 it stops due to a signal, you may want to use @code{handle}, or use
5532 @samp{signal 0} to resume execution (@pxref{Signals, ,Signals}),
5533 or you may step into the signal's handler (@pxref{stepping and signal
5538 @kindex c @r{(@code{continue})}
5539 @kindex fg @r{(resume foreground execution)}
5540 @item continue @r{[}@var{ignore-count}@r{]}
5541 @itemx c @r{[}@var{ignore-count}@r{]}
5542 @itemx fg @r{[}@var{ignore-count}@r{]}
5543 Resume program execution, at the address where your program last stopped;
5544 any breakpoints set at that address are bypassed. The optional argument
5545 @var{ignore-count} allows you to specify a further number of times to
5546 ignore a breakpoint at this location; its effect is like that of
5547 @code{ignore} (@pxref{Conditions, ,Break Conditions}).
5549 The argument @var{ignore-count} is meaningful only when your program
5550 stopped due to a breakpoint. At other times, the argument to
5551 @code{continue} is ignored.
5553 The synonyms @code{c} and @code{fg} (for @dfn{foreground}, as the
5554 debugged program is deemed to be the foreground program) are provided
5555 purely for convenience, and have exactly the same behavior as
5559 To resume execution at a different place, you can use @code{return}
5560 (@pxref{Returning, ,Returning from a Function}) to go back to the
5561 calling function; or @code{jump} (@pxref{Jumping, ,Continuing at a
5562 Different Address}) to go to an arbitrary location in your program.
5564 A typical technique for using stepping is to set a breakpoint
5565 (@pxref{Breakpoints, ,Breakpoints; Watchpoints; and Catchpoints}) at the
5566 beginning of the function or the section of your program where a problem
5567 is believed to lie, run your program until it stops at that breakpoint,
5568 and then step through the suspect area, examining the variables that are
5569 interesting, until you see the problem happen.
5573 @kindex s @r{(@code{step})}
5575 Continue running your program until control reaches a different source
5576 line, then stop it and return control to @value{GDBN}. This command is
5577 abbreviated @code{s}.
5580 @c "without debugging information" is imprecise; actually "without line
5581 @c numbers in the debugging information". (gcc -g1 has debugging info but
5582 @c not line numbers). But it seems complex to try to make that
5583 @c distinction here.
5584 @emph{Warning:} If you use the @code{step} command while control is
5585 within a function that was compiled without debugging information,
5586 execution proceeds until control reaches a function that does have
5587 debugging information. Likewise, it will not step into a function which
5588 is compiled without debugging information. To step through functions
5589 without debugging information, use the @code{stepi} command, described
5593 The @code{step} command only stops at the first instruction of a source
5594 line. This prevents the multiple stops that could otherwise occur in
5595 @code{switch} statements, @code{for} loops, etc. @code{step} continues
5596 to stop if a function that has debugging information is called within
5597 the line. In other words, @code{step} @emph{steps inside} any functions
5598 called within the line.
5600 Also, the @code{step} command only enters a function if there is line
5601 number information for the function. Otherwise it acts like the
5602 @code{next} command. This avoids problems when using @code{cc -gl}
5603 on @acronym{MIPS} machines. Previously, @code{step} entered subroutines if there
5604 was any debugging information about the routine.
5606 @item step @var{count}
5607 Continue running as in @code{step}, but do so @var{count} times. If a
5608 breakpoint is reached, or a signal not related to stepping occurs before
5609 @var{count} steps, stepping stops right away.
5612 @kindex n @r{(@code{next})}
5613 @item next @r{[}@var{count}@r{]}
5614 Continue to the next source line in the current (innermost) stack frame.
5615 This is similar to @code{step}, but function calls that appear within
5616 the line of code are executed without stopping. Execution stops when
5617 control reaches a different line of code at the original stack level
5618 that was executing when you gave the @code{next} command. This command
5619 is abbreviated @code{n}.
5621 An argument @var{count} is a repeat count, as for @code{step}.
5624 @c FIX ME!! Do we delete this, or is there a way it fits in with
5625 @c the following paragraph? --- Vctoria
5627 @c @code{next} within a function that lacks debugging information acts like
5628 @c @code{step}, but any function calls appearing within the code of the
5629 @c function are executed without stopping.
5631 The @code{next} command only stops at the first instruction of a
5632 source line. This prevents multiple stops that could otherwise occur in
5633 @code{switch} statements, @code{for} loops, etc.
5635 @kindex set step-mode
5637 @cindex functions without line info, and stepping
5638 @cindex stepping into functions with no line info
5639 @itemx set step-mode on
5640 The @code{set step-mode on} command causes the @code{step} command to
5641 stop at the first instruction of a function which contains no debug line
5642 information rather than stepping over it.
5644 This is useful in cases where you may be interested in inspecting the
5645 machine instructions of a function which has no symbolic info and do not
5646 want @value{GDBN} to automatically skip over this function.
5648 @item set step-mode off
5649 Causes the @code{step} command to step over any functions which contains no
5650 debug information. This is the default.
5652 @item show step-mode
5653 Show whether @value{GDBN} will stop in or step over functions without
5654 source line debug information.
5657 @kindex fin @r{(@code{finish})}
5659 Continue running until just after function in the selected stack frame
5660 returns. Print the returned value (if any). This command can be
5661 abbreviated as @code{fin}.
5663 Contrast this with the @code{return} command (@pxref{Returning,
5664 ,Returning from a Function}).
5666 @kindex set print finish
5667 @kindex show print finish
5668 @item set print finish @r{[}on|off@r{]}
5669 @itemx show print finish
5670 By default the @code{finish} command will show the value that is
5671 returned by the function. This can be disabled using @code{set print
5672 finish off}. When disabled, the value is still entered into the value
5673 history (@pxref{Value History}), but not displayed.
5676 @kindex u @r{(@code{until})}
5677 @cindex run until specified location
5680 Continue running until a source line past the current line, in the
5681 current stack frame, is reached. This command is used to avoid single
5682 stepping through a loop more than once. It is like the @code{next}
5683 command, except that when @code{until} encounters a jump, it
5684 automatically continues execution until the program counter is greater
5685 than the address of the jump.
5687 This means that when you reach the end of a loop after single stepping
5688 though it, @code{until} makes your program continue execution until it
5689 exits the loop. In contrast, a @code{next} command at the end of a loop
5690 simply steps back to the beginning of the loop, which forces you to step
5691 through the next iteration.
5693 @code{until} always stops your program if it attempts to exit the current
5696 @code{until} may produce somewhat counterintuitive results if the order
5697 of machine code does not match the order of the source lines. For
5698 example, in the following excerpt from a debugging session, the @code{f}
5699 (@code{frame}) command shows that execution is stopped at line
5700 @code{206}; yet when we use @code{until}, we get to line @code{195}:
5704 #0 main (argc=4, argv=0xf7fffae8) at m4.c:206
5706 (@value{GDBP}) until
5707 195 for ( ; argc > 0; NEXTARG) @{
5710 This happened because, for execution efficiency, the compiler had
5711 generated code for the loop closure test at the end, rather than the
5712 start, of the loop---even though the test in a C @code{for}-loop is
5713 written before the body of the loop. The @code{until} command appeared
5714 to step back to the beginning of the loop when it advanced to this
5715 expression; however, it has not really gone to an earlier
5716 statement---not in terms of the actual machine code.
5718 @code{until} with no argument works by means of single
5719 instruction stepping, and hence is slower than @code{until} with an
5722 @item until @var{location}
5723 @itemx u @var{location}
5724 Continue running your program until either the specified @var{location} is
5725 reached, or the current stack frame returns. The location is any of
5726 the forms described in @ref{Specify Location}.
5727 This form of the command uses temporary breakpoints, and
5728 hence is quicker than @code{until} without an argument. The specified
5729 location is actually reached only if it is in the current frame. This
5730 implies that @code{until} can be used to skip over recursive function
5731 invocations. For instance in the code below, if the current location is
5732 line @code{96}, issuing @code{until 99} will execute the program up to
5733 line @code{99} in the same invocation of factorial, i.e., after the inner
5734 invocations have returned.
5737 94 int factorial (int value)
5739 96 if (value > 1) @{
5740 97 value *= factorial (value - 1);
5747 @kindex advance @var{location}
5748 @item advance @var{location}
5749 Continue running the program up to the given @var{location}. An argument is
5750 required, which should be of one of the forms described in
5751 @ref{Specify Location}.
5752 Execution will also stop upon exit from the current stack
5753 frame. This command is similar to @code{until}, but @code{advance} will
5754 not skip over recursive function calls, and the target location doesn't
5755 have to be in the same frame as the current one.
5759 @kindex si @r{(@code{stepi})}
5761 @itemx stepi @var{arg}
5763 Execute one machine instruction, then stop and return to the debugger.
5765 It is often useful to do @samp{display/i $pc} when stepping by machine
5766 instructions. This makes @value{GDBN} automatically display the next
5767 instruction to be executed, each time your program stops. @xref{Auto
5768 Display,, Automatic Display}.
5770 An argument is a repeat count, as in @code{step}.
5774 @kindex ni @r{(@code{nexti})}
5776 @itemx nexti @var{arg}
5778 Execute one machine instruction, but if it is a function call,
5779 proceed until the function returns.
5781 An argument is a repeat count, as in @code{next}.
5785 @anchor{range stepping}
5786 @cindex range stepping
5787 @cindex target-assisted range stepping
5788 By default, and if available, @value{GDBN} makes use of
5789 target-assisted @dfn{range stepping}. In other words, whenever you
5790 use a stepping command (e.g., @code{step}, @code{next}), @value{GDBN}
5791 tells the target to step the corresponding range of instruction
5792 addresses instead of issuing multiple single-steps. This speeds up
5793 line stepping, particularly for remote targets. Ideally, there should
5794 be no reason you would want to turn range stepping off. However, it's
5795 possible that a bug in the debug info, a bug in the remote stub (for
5796 remote targets), or even a bug in @value{GDBN} could make line
5797 stepping behave incorrectly when target-assisted range stepping is
5798 enabled. You can use the following command to turn off range stepping
5802 @kindex set range-stepping
5803 @kindex show range-stepping
5804 @item set range-stepping
5805 @itemx show range-stepping
5806 Control whether range stepping is enabled.
5808 If @code{on}, and the target supports it, @value{GDBN} tells the
5809 target to step a range of addresses itself, instead of issuing
5810 multiple single-steps. If @code{off}, @value{GDBN} always issues
5811 single-steps, even if range stepping is supported by the target. The
5812 default is @code{on}.
5816 @node Skipping Over Functions and Files
5817 @section Skipping Over Functions and Files
5818 @cindex skipping over functions and files
5820 The program you are debugging may contain some functions which are
5821 uninteresting to debug. The @code{skip} command lets you tell @value{GDBN} to
5822 skip a function, all functions in a file or a particular function in
5823 a particular file when stepping.
5825 For example, consider the following C function:
5836 Suppose you wish to step into the functions @code{foo} and @code{bar}, but you
5837 are not interested in stepping through @code{boring}. If you run @code{step}
5838 at line 103, you'll enter @code{boring()}, but if you run @code{next}, you'll
5839 step over both @code{foo} and @code{boring}!
5841 One solution is to @code{step} into @code{boring} and use the @code{finish}
5842 command to immediately exit it. But this can become tedious if @code{boring}
5843 is called from many places.
5845 A more flexible solution is to execute @kbd{skip boring}. This instructs
5846 @value{GDBN} never to step into @code{boring}. Now when you execute
5847 @code{step} at line 103, you'll step over @code{boring} and directly into
5850 Functions may be skipped by providing either a function name, linespec
5851 (@pxref{Specify Location}), regular expression that matches the function's
5852 name, file name or a @code{glob}-style pattern that matches the file name.
5854 On Posix systems the form of the regular expression is
5855 ``Extended Regular Expressions''. See for example @samp{man 7 regex}
5856 on @sc{gnu}/Linux systems. On non-Posix systems the form of the regular
5857 expression is whatever is provided by the @code{regcomp} function of
5858 the underlying system.
5859 See for example @samp{man 7 glob} on @sc{gnu}/Linux systems for a
5860 description of @code{glob}-style patterns.
5864 @item skip @r{[}@var{options}@r{]}
5865 The basic form of the @code{skip} command takes zero or more options
5866 that specify what to skip.
5867 The @var{options} argument is any useful combination of the following:
5870 @item -file @var{file}
5871 @itemx -fi @var{file}
5872 Functions in @var{file} will be skipped over when stepping.
5874 @item -gfile @var{file-glob-pattern}
5875 @itemx -gfi @var{file-glob-pattern}
5876 @cindex skipping over files via glob-style patterns
5877 Functions in files matching @var{file-glob-pattern} will be skipped
5881 (gdb) skip -gfi utils/*.c
5884 @item -function @var{linespec}
5885 @itemx -fu @var{linespec}
5886 Functions named by @var{linespec} or the function containing the line
5887 named by @var{linespec} will be skipped over when stepping.
5888 @xref{Specify Location}.
5890 @item -rfunction @var{regexp}
5891 @itemx -rfu @var{regexp}
5892 @cindex skipping over functions via regular expressions
5893 Functions whose name matches @var{regexp} will be skipped over when stepping.
5895 This form is useful for complex function names.
5896 For example, there is generally no need to step into C@t{++} @code{std::string}
5897 constructors or destructors. Plus with C@t{++} templates it can be hard to
5898 write out the full name of the function, and often it doesn't matter what
5899 the template arguments are. Specifying the function to be skipped as a
5900 regular expression makes this easier.
5903 (gdb) skip -rfu ^std::(allocator|basic_string)<.*>::~?\1 *\(
5906 If you want to skip every templated C@t{++} constructor and destructor
5907 in the @code{std} namespace you can do:
5910 (gdb) skip -rfu ^std::([a-zA-z0-9_]+)<.*>::~?\1 *\(
5914 If no options are specified, the function you're currently debugging
5917 @kindex skip function
5918 @item skip function @r{[}@var{linespec}@r{]}
5919 After running this command, the function named by @var{linespec} or the
5920 function containing the line named by @var{linespec} will be skipped over when
5921 stepping. @xref{Specify Location}.
5923 If you do not specify @var{linespec}, the function you're currently debugging
5926 (If you have a function called @code{file} that you want to skip, use
5927 @kbd{skip function file}.)
5930 @item skip file @r{[}@var{filename}@r{]}
5931 After running this command, any function whose source lives in @var{filename}
5932 will be skipped over when stepping.
5935 (gdb) skip file boring.c
5936 File boring.c will be skipped when stepping.
5939 If you do not specify @var{filename}, functions whose source lives in the file
5940 you're currently debugging will be skipped.
5943 Skips can be listed, deleted, disabled, and enabled, much like breakpoints.
5944 These are the commands for managing your list of skips:
5948 @item info skip @r{[}@var{range}@r{]}
5949 Print details about the specified skip(s). If @var{range} is not specified,
5950 print a table with details about all functions and files marked for skipping.
5951 @code{info skip} prints the following information about each skip:
5955 A number identifying this skip.
5956 @item Enabled or Disabled
5957 Enabled skips are marked with @samp{y}.
5958 Disabled skips are marked with @samp{n}.
5960 If the file name is a @samp{glob} pattern this is @samp{y}.
5961 Otherwise it is @samp{n}.
5963 The name or @samp{glob} pattern of the file to be skipped.
5964 If no file is specified this is @samp{<none>}.
5966 If the function name is a @samp{regular expression} this is @samp{y}.
5967 Otherwise it is @samp{n}.
5969 The name or regular expression of the function to skip.
5970 If no function is specified this is @samp{<none>}.
5974 @item skip delete @r{[}@var{range}@r{]}
5975 Delete the specified skip(s). If @var{range} is not specified, delete all
5979 @item skip enable @r{[}@var{range}@r{]}
5980 Enable the specified skip(s). If @var{range} is not specified, enable all
5983 @kindex skip disable
5984 @item skip disable @r{[}@var{range}@r{]}
5985 Disable the specified skip(s). If @var{range} is not specified, disable all
5988 @kindex set debug skip
5989 @item set debug skip @r{[}on|off@r{]}
5990 Set whether to print the debug output about skipping files and functions.
5992 @kindex show debug skip
5993 @item show debug skip
5994 Show whether the debug output about skipping files and functions is printed.
6002 A signal is an asynchronous event that can happen in a program. The
6003 operating system defines the possible kinds of signals, and gives each
6004 kind a name and a number. For example, in Unix @code{SIGINT} is the
6005 signal a program gets when you type an interrupt character (often @kbd{Ctrl-c});
6006 @code{SIGSEGV} is the signal a program gets from referencing a place in
6007 memory far away from all the areas in use; @code{SIGALRM} occurs when
6008 the alarm clock timer goes off (which happens only if your program has
6009 requested an alarm).
6011 @cindex fatal signals
6012 Some signals, including @code{SIGALRM}, are a normal part of the
6013 functioning of your program. Others, such as @code{SIGSEGV}, indicate
6014 errors; these signals are @dfn{fatal} (they kill your program immediately) if the
6015 program has not specified in advance some other way to handle the signal.
6016 @code{SIGINT} does not indicate an error in your program, but it is normally
6017 fatal so it can carry out the purpose of the interrupt: to kill the program.
6019 @value{GDBN} has the ability to detect any occurrence of a signal in your
6020 program. You can tell @value{GDBN} in advance what to do for each kind of
6023 @cindex handling signals
6024 Normally, @value{GDBN} is set up to let the non-erroneous signals like
6025 @code{SIGALRM} be silently passed to your program
6026 (so as not to interfere with their role in the program's functioning)
6027 but to stop your program immediately whenever an error signal happens.
6028 You can change these settings with the @code{handle} command.
6031 @kindex info signals
6035 Print a table of all the kinds of signals and how @value{GDBN} has been told to
6036 handle each one. You can use this to see the signal numbers of all
6037 the defined types of signals.
6039 @item info signals @var{sig}
6040 Similar, but print information only about the specified signal number.
6042 @code{info handle} is an alias for @code{info signals}.
6044 @item catch signal @r{[}@var{signal}@dots{} @r{|} @samp{all}@r{]}
6045 Set a catchpoint for the indicated signals. @xref{Set Catchpoints},
6046 for details about this command.
6049 @item handle @var{signal} @r{[}@var{keywords}@dots{}@r{]}
6050 Change the way @value{GDBN} handles signal @var{signal}. The @var{signal}
6051 can be the number of a signal or its name (with or without the
6052 @samp{SIG} at the beginning); a list of signal numbers of the form
6053 @samp{@var{low}-@var{high}}; or the word @samp{all}, meaning all the
6054 known signals. Optional arguments @var{keywords}, described below,
6055 say what change to make.
6059 The keywords allowed by the @code{handle} command can be abbreviated.
6060 Their full names are:
6064 @value{GDBN} should not stop your program when this signal happens. It may
6065 still print a message telling you that the signal has come in.
6068 @value{GDBN} should stop your program when this signal happens. This implies
6069 the @code{print} keyword as well.
6072 @value{GDBN} should print a message when this signal happens.
6075 @value{GDBN} should not mention the occurrence of the signal at all. This
6076 implies the @code{nostop} keyword as well.
6080 @value{GDBN} should allow your program to see this signal; your program
6081 can handle the signal, or else it may terminate if the signal is fatal
6082 and not handled. @code{pass} and @code{noignore} are synonyms.
6086 @value{GDBN} should not allow your program to see this signal.
6087 @code{nopass} and @code{ignore} are synonyms.
6091 When a signal stops your program, the signal is not visible to the
6093 continue. Your program sees the signal then, if @code{pass} is in
6094 effect for the signal in question @emph{at that time}. In other words,
6095 after @value{GDBN} reports a signal, you can use the @code{handle}
6096 command with @code{pass} or @code{nopass} to control whether your
6097 program sees that signal when you continue.
6099 The default is set to @code{nostop}, @code{noprint}, @code{pass} for
6100 non-erroneous signals such as @code{SIGALRM}, @code{SIGWINCH} and
6101 @code{SIGCHLD}, and to @code{stop}, @code{print}, @code{pass} for the
6104 You can also use the @code{signal} command to prevent your program from
6105 seeing a signal, or cause it to see a signal it normally would not see,
6106 or to give it any signal at any time. For example, if your program stopped
6107 due to some sort of memory reference error, you might store correct
6108 values into the erroneous variables and continue, hoping to see more
6109 execution; but your program would probably terminate immediately as
6110 a result of the fatal signal once it saw the signal. To prevent this,
6111 you can continue with @samp{signal 0}. @xref{Signaling, ,Giving your
6114 @cindex stepping and signal handlers
6115 @anchor{stepping and signal handlers}
6117 @value{GDBN} optimizes for stepping the mainline code. If a signal
6118 that has @code{handle nostop} and @code{handle pass} set arrives while
6119 a stepping command (e.g., @code{stepi}, @code{step}, @code{next}) is
6120 in progress, @value{GDBN} lets the signal handler run and then resumes
6121 stepping the mainline code once the signal handler returns. In other
6122 words, @value{GDBN} steps over the signal handler. This prevents
6123 signals that you've specified as not interesting (with @code{handle
6124 nostop}) from changing the focus of debugging unexpectedly. Note that
6125 the signal handler itself may still hit a breakpoint, stop for another
6126 signal that has @code{handle stop} in effect, or for any other event
6127 that normally results in stopping the stepping command sooner. Also
6128 note that @value{GDBN} still informs you that the program received a
6129 signal if @code{handle print} is set.
6131 @anchor{stepping into signal handlers}
6133 If you set @code{handle pass} for a signal, and your program sets up a
6134 handler for it, then issuing a stepping command, such as @code{step}
6135 or @code{stepi}, when your program is stopped due to the signal will
6136 step @emph{into} the signal handler (if the target supports that).
6138 Likewise, if you use the @code{queue-signal} command to queue a signal
6139 to be delivered to the current thread when execution of the thread
6140 resumes (@pxref{Signaling, ,Giving your Program a Signal}), then a
6141 stepping command will step into the signal handler.
6143 Here's an example, using @code{stepi} to step to the first instruction
6144 of @code{SIGUSR1}'s handler:
6147 (@value{GDBP}) handle SIGUSR1
6148 Signal Stop Print Pass to program Description
6149 SIGUSR1 Yes Yes Yes User defined signal 1
6153 Program received signal SIGUSR1, User defined signal 1.
6154 main () sigusr1.c:28
6157 sigusr1_handler () at sigusr1.c:9
6161 The same, but using @code{queue-signal} instead of waiting for the
6162 program to receive the signal first:
6167 (@value{GDBP}) queue-signal SIGUSR1
6169 sigusr1_handler () at sigusr1.c:9
6174 @cindex extra signal information
6175 @anchor{extra signal information}
6177 On some targets, @value{GDBN} can inspect extra signal information
6178 associated with the intercepted signal, before it is actually
6179 delivered to the program being debugged. This information is exported
6180 by the convenience variable @code{$_siginfo}, and consists of data
6181 that is passed by the kernel to the signal handler at the time of the
6182 receipt of a signal. The data type of the information itself is
6183 target dependent. You can see the data type using the @code{ptype
6184 $_siginfo} command. On Unix systems, it typically corresponds to the
6185 standard @code{siginfo_t} type, as defined in the @file{signal.h}
6188 Here's an example, on a @sc{gnu}/Linux system, printing the stray
6189 referenced address that raised a segmentation fault.
6193 (@value{GDBP}) continue
6194 Program received signal SIGSEGV, Segmentation fault.
6195 0x0000000000400766 in main ()
6197 (@value{GDBP}) ptype $_siginfo
6204 struct @{...@} _kill;
6205 struct @{...@} _timer;
6207 struct @{...@} _sigchld;
6208 struct @{...@} _sigfault;
6209 struct @{...@} _sigpoll;
6212 (@value{GDBP}) ptype $_siginfo._sifields._sigfault
6216 (@value{GDBP}) p $_siginfo._sifields._sigfault.si_addr
6217 $1 = (void *) 0x7ffff7ff7000
6221 Depending on target support, @code{$_siginfo} may also be writable.
6223 @cindex Intel MPX boundary violations
6224 @cindex boundary violations, Intel MPX
6225 On some targets, a @code{SIGSEGV} can be caused by a boundary
6226 violation, i.e., accessing an address outside of the allowed range.
6227 In those cases @value{GDBN} may displays additional information,
6228 depending on how @value{GDBN} has been told to handle the signal.
6229 With @code{handle stop SIGSEGV}, @value{GDBN} displays the violation
6230 kind: "Upper" or "Lower", the memory address accessed and the
6231 bounds, while with @code{handle nostop SIGSEGV} no additional
6232 information is displayed.
6234 The usual output of a segfault is:
6236 Program received signal SIGSEGV, Segmentation fault
6237 0x0000000000400d7c in upper () at i386-mpx-sigsegv.c:68
6238 68 value = *(p + len);
6241 While a bound violation is presented as:
6243 Program received signal SIGSEGV, Segmentation fault
6244 Upper bound violation while accessing address 0x7fffffffc3b3
6245 Bounds: [lower = 0x7fffffffc390, upper = 0x7fffffffc3a3]
6246 0x0000000000400d7c in upper () at i386-mpx-sigsegv.c:68
6247 68 value = *(p + len);
6251 @section Stopping and Starting Multi-thread Programs
6253 @cindex stopped threads
6254 @cindex threads, stopped
6256 @cindex continuing threads
6257 @cindex threads, continuing
6259 @value{GDBN} supports debugging programs with multiple threads
6260 (@pxref{Threads,, Debugging Programs with Multiple Threads}). There
6261 are two modes of controlling execution of your program within the
6262 debugger. In the default mode, referred to as @dfn{all-stop mode},
6263 when any thread in your program stops (for example, at a breakpoint
6264 or while being stepped), all other threads in the program are also stopped by
6265 @value{GDBN}. On some targets, @value{GDBN} also supports
6266 @dfn{non-stop mode}, in which other threads can continue to run freely while
6267 you examine the stopped thread in the debugger.
6270 * All-Stop Mode:: All threads stop when GDB takes control
6271 * Non-Stop Mode:: Other threads continue to execute
6272 * Background Execution:: Running your program asynchronously
6273 * Thread-Specific Breakpoints:: Controlling breakpoints
6274 * Interrupted System Calls:: GDB may interfere with system calls
6275 * Observer Mode:: GDB does not alter program behavior
6279 @subsection All-Stop Mode
6281 @cindex all-stop mode
6283 In all-stop mode, whenever your program stops under @value{GDBN} for any reason,
6284 @emph{all} threads of execution stop, not just the current thread. This
6285 allows you to examine the overall state of the program, including
6286 switching between threads, without worrying that things may change
6289 Conversely, whenever you restart the program, @emph{all} threads start
6290 executing. @emph{This is true even when single-stepping} with commands
6291 like @code{step} or @code{next}.
6293 In particular, @value{GDBN} cannot single-step all threads in lockstep.
6294 Since thread scheduling is up to your debugging target's operating
6295 system (not controlled by @value{GDBN}), other threads may
6296 execute more than one statement while the current thread completes a
6297 single step. Moreover, in general other threads stop in the middle of a
6298 statement, rather than at a clean statement boundary, when the program
6301 You might even find your program stopped in another thread after
6302 continuing or even single-stepping. This happens whenever some other
6303 thread runs into a breakpoint, a signal, or an exception before the
6304 first thread completes whatever you requested.
6306 @cindex automatic thread selection
6307 @cindex switching threads automatically
6308 @cindex threads, automatic switching
6309 Whenever @value{GDBN} stops your program, due to a breakpoint or a
6310 signal, it automatically selects the thread where that breakpoint or
6311 signal happened. @value{GDBN} alerts you to the context switch with a
6312 message such as @samp{[Switching to Thread @var{n}]} to identify the
6315 On some OSes, you can modify @value{GDBN}'s default behavior by
6316 locking the OS scheduler to allow only a single thread to run.
6319 @item set scheduler-locking @var{mode}
6320 @cindex scheduler locking mode
6321 @cindex lock scheduler
6322 Set the scheduler locking mode. It applies to normal execution,
6323 record mode, and replay mode. If it is @code{off}, then there is no
6324 locking and any thread may run at any time. If @code{on}, then only
6325 the current thread may run when the inferior is resumed. The
6326 @code{step} mode optimizes for single-stepping; it prevents other
6327 threads from preempting the current thread while you are stepping, so
6328 that the focus of debugging does not change unexpectedly. Other
6329 threads never get a chance to run when you step, and they are
6330 completely free to run when you use commands like @samp{continue},
6331 @samp{until}, or @samp{finish}. However, unless another thread hits a
6332 breakpoint during its timeslice, @value{GDBN} does not change the
6333 current thread away from the thread that you are debugging. The
6334 @code{replay} mode behaves like @code{off} in record mode and like
6335 @code{on} in replay mode.
6337 @item show scheduler-locking
6338 Display the current scheduler locking mode.
6341 @cindex resume threads of multiple processes simultaneously
6342 By default, when you issue one of the execution commands such as
6343 @code{continue}, @code{next} or @code{step}, @value{GDBN} allows only
6344 threads of the current inferior to run. For example, if @value{GDBN}
6345 is attached to two inferiors, each with two threads, the
6346 @code{continue} command resumes only the two threads of the current
6347 inferior. This is useful, for example, when you debug a program that
6348 forks and you want to hold the parent stopped (so that, for instance,
6349 it doesn't run to exit), while you debug the child. In other
6350 situations, you may not be interested in inspecting the current state
6351 of any of the processes @value{GDBN} is attached to, and you may want
6352 to resume them all until some breakpoint is hit. In the latter case,
6353 you can instruct @value{GDBN} to allow all threads of all the
6354 inferiors to run with the @w{@code{set schedule-multiple}} command.
6357 @kindex set schedule-multiple
6358 @item set schedule-multiple
6359 Set the mode for allowing threads of multiple processes to be resumed
6360 when an execution command is issued. When @code{on}, all threads of
6361 all processes are allowed to run. When @code{off}, only the threads
6362 of the current process are resumed. The default is @code{off}. The
6363 @code{scheduler-locking} mode takes precedence when set to @code{on},
6364 or while you are stepping and set to @code{step}.
6366 @item show schedule-multiple
6367 Display the current mode for resuming the execution of threads of
6372 @subsection Non-Stop Mode
6374 @cindex non-stop mode
6376 @c This section is really only a place-holder, and needs to be expanded
6377 @c with more details.
6379 For some multi-threaded targets, @value{GDBN} supports an optional
6380 mode of operation in which you can examine stopped program threads in
6381 the debugger while other threads continue to execute freely. This
6382 minimizes intrusion when debugging live systems, such as programs
6383 where some threads have real-time constraints or must continue to
6384 respond to external events. This is referred to as @dfn{non-stop} mode.
6386 In non-stop mode, when a thread stops to report a debugging event,
6387 @emph{only} that thread is stopped; @value{GDBN} does not stop other
6388 threads as well, in contrast to the all-stop mode behavior. Additionally,
6389 execution commands such as @code{continue} and @code{step} apply by default
6390 only to the current thread in non-stop mode, rather than all threads as
6391 in all-stop mode. This allows you to control threads explicitly in
6392 ways that are not possible in all-stop mode --- for example, stepping
6393 one thread while allowing others to run freely, stepping
6394 one thread while holding all others stopped, or stepping several threads
6395 independently and simultaneously.
6397 To enter non-stop mode, use this sequence of commands before you run
6398 or attach to your program:
6401 # If using the CLI, pagination breaks non-stop.
6404 # Finally, turn it on!
6408 You can use these commands to manipulate the non-stop mode setting:
6411 @kindex set non-stop
6412 @item set non-stop on
6413 Enable selection of non-stop mode.
6414 @item set non-stop off
6415 Disable selection of non-stop mode.
6416 @kindex show non-stop
6418 Show the current non-stop enablement setting.
6421 Note these commands only reflect whether non-stop mode is enabled,
6422 not whether the currently-executing program is being run in non-stop mode.
6423 In particular, the @code{set non-stop} preference is only consulted when
6424 @value{GDBN} starts or connects to the target program, and it is generally
6425 not possible to switch modes once debugging has started. Furthermore,
6426 since not all targets support non-stop mode, even when you have enabled
6427 non-stop mode, @value{GDBN} may still fall back to all-stop operation by
6430 In non-stop mode, all execution commands apply only to the current thread
6431 by default. That is, @code{continue} only continues one thread.
6432 To continue all threads, issue @code{continue -a} or @code{c -a}.
6434 You can use @value{GDBN}'s background execution commands
6435 (@pxref{Background Execution}) to run some threads in the background
6436 while you continue to examine or step others from @value{GDBN}.
6437 The MI execution commands (@pxref{GDB/MI Program Execution}) are
6438 always executed asynchronously in non-stop mode.
6440 Suspending execution is done with the @code{interrupt} command when
6441 running in the background, or @kbd{Ctrl-c} during foreground execution.
6442 In all-stop mode, this stops the whole process;
6443 but in non-stop mode the interrupt applies only to the current thread.
6444 To stop the whole program, use @code{interrupt -a}.
6446 Other execution commands do not currently support the @code{-a} option.
6448 In non-stop mode, when a thread stops, @value{GDBN} doesn't automatically make
6449 that thread current, as it does in all-stop mode. This is because the
6450 thread stop notifications are asynchronous with respect to @value{GDBN}'s
6451 command interpreter, and it would be confusing if @value{GDBN} unexpectedly
6452 changed to a different thread just as you entered a command to operate on the
6453 previously current thread.
6455 @node Background Execution
6456 @subsection Background Execution
6458 @cindex foreground execution
6459 @cindex background execution
6460 @cindex asynchronous execution
6461 @cindex execution, foreground, background and asynchronous
6463 @value{GDBN}'s execution commands have two variants: the normal
6464 foreground (synchronous) behavior, and a background
6465 (asynchronous) behavior. In foreground execution, @value{GDBN} waits for
6466 the program to report that some thread has stopped before prompting for
6467 another command. In background execution, @value{GDBN} immediately gives
6468 a command prompt so that you can issue other commands while your program runs.
6470 If the target doesn't support async mode, @value{GDBN} issues an error
6471 message if you attempt to use the background execution commands.
6473 @cindex @code{&}, background execution of commands
6474 To specify background execution, add a @code{&} to the command. For example,
6475 the background form of the @code{continue} command is @code{continue&}, or
6476 just @code{c&}. The execution commands that accept background execution
6482 @xref{Starting, , Starting your Program}.
6486 @xref{Attach, , Debugging an Already-running Process}.
6490 @xref{Continuing and Stepping, step}.
6494 @xref{Continuing and Stepping, stepi}.
6498 @xref{Continuing and Stepping, next}.
6502 @xref{Continuing and Stepping, nexti}.
6506 @xref{Continuing and Stepping, continue}.
6510 @xref{Continuing and Stepping, finish}.
6514 @xref{Continuing and Stepping, until}.
6518 Background execution is especially useful in conjunction with non-stop
6519 mode for debugging programs with multiple threads; see @ref{Non-Stop Mode}.
6520 However, you can also use these commands in the normal all-stop mode with
6521 the restriction that you cannot issue another execution command until the
6522 previous one finishes. Examples of commands that are valid in all-stop
6523 mode while the program is running include @code{help} and @code{info break}.
6525 You can interrupt your program while it is running in the background by
6526 using the @code{interrupt} command.
6533 Suspend execution of the running program. In all-stop mode,
6534 @code{interrupt} stops the whole process, but in non-stop mode, it stops
6535 only the current thread. To stop the whole program in non-stop mode,
6536 use @code{interrupt -a}.
6539 @node Thread-Specific Breakpoints
6540 @subsection Thread-Specific Breakpoints
6542 When your program has multiple threads (@pxref{Threads,, Debugging
6543 Programs with Multiple Threads}), you can choose whether to set
6544 breakpoints on all threads, or on a particular thread.
6547 @cindex breakpoints and threads
6548 @cindex thread breakpoints
6549 @kindex break @dots{} thread @var{thread-id}
6550 @item break @var{location} thread @var{thread-id}
6551 @itemx break @var{location} thread @var{thread-id} if @dots{}
6552 @var{location} specifies source lines; there are several ways of
6553 writing them (@pxref{Specify Location}), but the effect is always to
6554 specify some source line.
6556 Use the qualifier @samp{thread @var{thread-id}} with a breakpoint command
6557 to specify that you only want @value{GDBN} to stop the program when a
6558 particular thread reaches this breakpoint. The @var{thread-id} specifier
6559 is one of the thread identifiers assigned by @value{GDBN}, shown
6560 in the first column of the @samp{info threads} display.
6562 If you do not specify @samp{thread @var{thread-id}} when you set a
6563 breakpoint, the breakpoint applies to @emph{all} threads of your
6566 You can use the @code{thread} qualifier on conditional breakpoints as
6567 well; in this case, place @samp{thread @var{thread-id}} before or
6568 after the breakpoint condition, like this:
6571 (@value{GDBP}) break frik.c:13 thread 28 if bartab > lim
6576 Thread-specific breakpoints are automatically deleted when
6577 @value{GDBN} detects the corresponding thread is no longer in the
6578 thread list. For example:
6582 Thread-specific breakpoint 3 deleted - thread 28 no longer in the thread list.
6585 There are several ways for a thread to disappear, such as a regular
6586 thread exit, but also when you detach from the process with the
6587 @code{detach} command (@pxref{Attach, ,Debugging an Already-running
6588 Process}), or if @value{GDBN} loses the remote connection
6589 (@pxref{Remote Debugging}), etc. Note that with some targets,
6590 @value{GDBN} is only able to detect a thread has exited when the user
6591 explictly asks for the thread list with the @code{info threads}
6594 @node Interrupted System Calls
6595 @subsection Interrupted System Calls
6597 @cindex thread breakpoints and system calls
6598 @cindex system calls and thread breakpoints
6599 @cindex premature return from system calls
6600 There is an unfortunate side effect when using @value{GDBN} to debug
6601 multi-threaded programs. If one thread stops for a
6602 breakpoint, or for some other reason, and another thread is blocked in a
6603 system call, then the system call may return prematurely. This is a
6604 consequence of the interaction between multiple threads and the signals
6605 that @value{GDBN} uses to implement breakpoints and other events that
6608 To handle this problem, your program should check the return value of
6609 each system call and react appropriately. This is good programming
6612 For example, do not write code like this:
6618 The call to @code{sleep} will return early if a different thread stops
6619 at a breakpoint or for some other reason.
6621 Instead, write this:
6626 unslept = sleep (unslept);
6629 A system call is allowed to return early, so the system is still
6630 conforming to its specification. But @value{GDBN} does cause your
6631 multi-threaded program to behave differently than it would without
6634 Also, @value{GDBN} uses internal breakpoints in the thread library to
6635 monitor certain events such as thread creation and thread destruction.
6636 When such an event happens, a system call in another thread may return
6637 prematurely, even though your program does not appear to stop.
6640 @subsection Observer Mode
6642 If you want to build on non-stop mode and observe program behavior
6643 without any chance of disruption by @value{GDBN}, you can set
6644 variables to disable all of the debugger's attempts to modify state,
6645 whether by writing memory, inserting breakpoints, etc. These operate
6646 at a low level, intercepting operations from all commands.
6648 When all of these are set to @code{off}, then @value{GDBN} is said to
6649 be @dfn{observer mode}. As a convenience, the variable
6650 @code{observer} can be set to disable these, plus enable non-stop
6653 Note that @value{GDBN} will not prevent you from making nonsensical
6654 combinations of these settings. For instance, if you have enabled
6655 @code{may-insert-breakpoints} but disabled @code{may-write-memory},
6656 then breakpoints that work by writing trap instructions into the code
6657 stream will still not be able to be placed.
6662 @item set observer on
6663 @itemx set observer off
6664 When set to @code{on}, this disables all the permission variables
6665 below (except for @code{insert-fast-tracepoints}), plus enables
6666 non-stop debugging. Setting this to @code{off} switches back to
6667 normal debugging, though remaining in non-stop mode.
6670 Show whether observer mode is on or off.
6672 @kindex may-write-registers
6673 @item set may-write-registers on
6674 @itemx set may-write-registers off
6675 This controls whether @value{GDBN} will attempt to alter the values of
6676 registers, such as with assignment expressions in @code{print}, or the
6677 @code{jump} command. It defaults to @code{on}.
6679 @item show may-write-registers
6680 Show the current permission to write registers.
6682 @kindex may-write-memory
6683 @item set may-write-memory on
6684 @itemx set may-write-memory off
6685 This controls whether @value{GDBN} will attempt to alter the contents
6686 of memory, such as with assignment expressions in @code{print}. It
6687 defaults to @code{on}.
6689 @item show may-write-memory
6690 Show the current permission to write memory.
6692 @kindex may-insert-breakpoints
6693 @item set may-insert-breakpoints on
6694 @itemx set may-insert-breakpoints off
6695 This controls whether @value{GDBN} will attempt to insert breakpoints.
6696 This affects all breakpoints, including internal breakpoints defined
6697 by @value{GDBN}. It defaults to @code{on}.
6699 @item show may-insert-breakpoints
6700 Show the current permission to insert breakpoints.
6702 @kindex may-insert-tracepoints
6703 @item set may-insert-tracepoints on
6704 @itemx set may-insert-tracepoints off
6705 This controls whether @value{GDBN} will attempt to insert (regular)
6706 tracepoints at the beginning of a tracing experiment. It affects only
6707 non-fast tracepoints, fast tracepoints being under the control of
6708 @code{may-insert-fast-tracepoints}. It defaults to @code{on}.
6710 @item show may-insert-tracepoints
6711 Show the current permission to insert tracepoints.
6713 @kindex may-insert-fast-tracepoints
6714 @item set may-insert-fast-tracepoints on
6715 @itemx set may-insert-fast-tracepoints off
6716 This controls whether @value{GDBN} will attempt to insert fast
6717 tracepoints at the beginning of a tracing experiment. It affects only
6718 fast tracepoints, regular (non-fast) tracepoints being under the
6719 control of @code{may-insert-tracepoints}. It defaults to @code{on}.
6721 @item show may-insert-fast-tracepoints
6722 Show the current permission to insert fast tracepoints.
6724 @kindex may-interrupt
6725 @item set may-interrupt on
6726 @itemx set may-interrupt off
6727 This controls whether @value{GDBN} will attempt to interrupt or stop
6728 program execution. When this variable is @code{off}, the
6729 @code{interrupt} command will have no effect, nor will
6730 @kbd{Ctrl-c}. It defaults to @code{on}.
6732 @item show may-interrupt
6733 Show the current permission to interrupt or stop the program.
6737 @node Reverse Execution
6738 @chapter Running programs backward
6739 @cindex reverse execution
6740 @cindex running programs backward
6742 When you are debugging a program, it is not unusual to realize that
6743 you have gone too far, and some event of interest has already happened.
6744 If the target environment supports it, @value{GDBN} can allow you to
6745 ``rewind'' the program by running it backward.
6747 A target environment that supports reverse execution should be able
6748 to ``undo'' the changes in machine state that have taken place as the
6749 program was executing normally. Variables, registers etc.@: should
6750 revert to their previous values. Obviously this requires a great
6751 deal of sophistication on the part of the target environment; not
6752 all target environments can support reverse execution.
6754 When a program is executed in reverse, the instructions that
6755 have most recently been executed are ``un-executed'', in reverse
6756 order. The program counter runs backward, following the previous
6757 thread of execution in reverse. As each instruction is ``un-executed'',
6758 the values of memory and/or registers that were changed by that
6759 instruction are reverted to their previous states. After executing
6760 a piece of source code in reverse, all side effects of that code
6761 should be ``undone'', and all variables should be returned to their
6762 prior values@footnote{
6763 Note that some side effects are easier to undo than others. For instance,
6764 memory and registers are relatively easy, but device I/O is hard. Some
6765 targets may be able undo things like device I/O, and some may not.
6767 The contract between @value{GDBN} and the reverse executing target
6768 requires only that the target do something reasonable when
6769 @value{GDBN} tells it to execute backwards, and then report the
6770 results back to @value{GDBN}. Whatever the target reports back to
6771 @value{GDBN}, @value{GDBN} will report back to the user. @value{GDBN}
6772 assumes that the memory and registers that the target reports are in a
6773 consistant state, but @value{GDBN} accepts whatever it is given.
6776 On some platforms, @value{GDBN} has built-in support for reverse
6777 execution, activated with the @code{record} or @code{record btrace}
6778 commands. @xref{Process Record and Replay}. Some remote targets,
6779 typically full system emulators, support reverse execution directly
6780 without requiring any special command.
6782 If you are debugging in a target environment that supports
6783 reverse execution, @value{GDBN} provides the following commands.
6786 @kindex reverse-continue
6787 @kindex rc @r{(@code{reverse-continue})}
6788 @item reverse-continue @r{[}@var{ignore-count}@r{]}
6789 @itemx rc @r{[}@var{ignore-count}@r{]}
6790 Beginning at the point where your program last stopped, start executing
6791 in reverse. Reverse execution will stop for breakpoints and synchronous
6792 exceptions (signals), just like normal execution. Behavior of
6793 asynchronous signals depends on the target environment.
6795 @kindex reverse-step
6796 @kindex rs @r{(@code{step})}
6797 @item reverse-step @r{[}@var{count}@r{]}
6798 Run the program backward until control reaches the start of a
6799 different source line; then stop it, and return control to @value{GDBN}.
6801 Like the @code{step} command, @code{reverse-step} will only stop
6802 at the beginning of a source line. It ``un-executes'' the previously
6803 executed source line. If the previous source line included calls to
6804 debuggable functions, @code{reverse-step} will step (backward) into
6805 the called function, stopping at the beginning of the @emph{last}
6806 statement in the called function (typically a return statement).
6808 Also, as with the @code{step} command, if non-debuggable functions are
6809 called, @code{reverse-step} will run thru them backward without stopping.
6811 @kindex reverse-stepi
6812 @kindex rsi @r{(@code{reverse-stepi})}
6813 @item reverse-stepi @r{[}@var{count}@r{]}
6814 Reverse-execute one machine instruction. Note that the instruction
6815 to be reverse-executed is @emph{not} the one pointed to by the program
6816 counter, but the instruction executed prior to that one. For instance,
6817 if the last instruction was a jump, @code{reverse-stepi} will take you
6818 back from the destination of the jump to the jump instruction itself.
6820 @kindex reverse-next
6821 @kindex rn @r{(@code{reverse-next})}
6822 @item reverse-next @r{[}@var{count}@r{]}
6823 Run backward to the beginning of the previous line executed in
6824 the current (innermost) stack frame. If the line contains function
6825 calls, they will be ``un-executed'' without stopping. Starting from
6826 the first line of a function, @code{reverse-next} will take you back
6827 to the caller of that function, @emph{before} the function was called,
6828 just as the normal @code{next} command would take you from the last
6829 line of a function back to its return to its caller
6830 @footnote{Unless the code is too heavily optimized.}.
6832 @kindex reverse-nexti
6833 @kindex rni @r{(@code{reverse-nexti})}
6834 @item reverse-nexti @r{[}@var{count}@r{]}
6835 Like @code{nexti}, @code{reverse-nexti} executes a single instruction
6836 in reverse, except that called functions are ``un-executed'' atomically.
6837 That is, if the previously executed instruction was a return from
6838 another function, @code{reverse-nexti} will continue to execute
6839 in reverse until the call to that function (from the current stack
6842 @kindex reverse-finish
6843 @item reverse-finish
6844 Just as the @code{finish} command takes you to the point where the
6845 current function returns, @code{reverse-finish} takes you to the point
6846 where it was called. Instead of ending up at the end of the current
6847 function invocation, you end up at the beginning.
6849 @kindex set exec-direction
6850 @item set exec-direction
6851 Set the direction of target execution.
6852 @item set exec-direction reverse
6853 @cindex execute forward or backward in time
6854 @value{GDBN} will perform all execution commands in reverse, until the
6855 exec-direction mode is changed to ``forward''. Affected commands include
6856 @code{step, stepi, next, nexti, continue, and finish}. The @code{return}
6857 command cannot be used in reverse mode.
6858 @item set exec-direction forward
6859 @value{GDBN} will perform all execution commands in the normal fashion.
6860 This is the default.
6864 @node Process Record and Replay
6865 @chapter Recording Inferior's Execution and Replaying It
6866 @cindex process record and replay
6867 @cindex recording inferior's execution and replaying it
6869 On some platforms, @value{GDBN} provides a special @dfn{process record
6870 and replay} target that can record a log of the process execution, and
6871 replay it later with both forward and reverse execution commands.
6874 When this target is in use, if the execution log includes the record
6875 for the next instruction, @value{GDBN} will debug in @dfn{replay
6876 mode}. In the replay mode, the inferior does not really execute code
6877 instructions. Instead, all the events that normally happen during
6878 code execution are taken from the execution log. While code is not
6879 really executed in replay mode, the values of registers (including the
6880 program counter register) and the memory of the inferior are still
6881 changed as they normally would. Their contents are taken from the
6885 If the record for the next instruction is not in the execution log,
6886 @value{GDBN} will debug in @dfn{record mode}. In this mode, the
6887 inferior executes normally, and @value{GDBN} records the execution log
6890 The process record and replay target supports reverse execution
6891 (@pxref{Reverse Execution}), even if the platform on which the
6892 inferior runs does not. However, the reverse execution is limited in
6893 this case by the range of the instructions recorded in the execution
6894 log. In other words, reverse execution on platforms that don't
6895 support it directly can only be done in the replay mode.
6897 When debugging in the reverse direction, @value{GDBN} will work in
6898 replay mode as long as the execution log includes the record for the
6899 previous instruction; otherwise, it will work in record mode, if the
6900 platform supports reverse execution, or stop if not.
6902 Currently, process record and replay is supported on ARM, Aarch64,
6903 Moxie, PowerPC, PowerPC64, S/390, and x86 (i386/amd64) running
6904 GNU/Linux. Process record and replay can be used both when native
6905 debugging, and when remote debugging via @code{gdbserver}.
6907 For architecture environments that support process record and replay,
6908 @value{GDBN} provides the following commands:
6911 @kindex target record
6912 @kindex target record-full
6913 @kindex target record-btrace
6916 @kindex record btrace
6917 @kindex record btrace bts
6918 @kindex record btrace pt
6924 @kindex rec btrace bts
6925 @kindex rec btrace pt
6928 @item record @var{method}
6929 This command starts the process record and replay target. The
6930 recording method can be specified as parameter. Without a parameter
6931 the command uses the @code{full} recording method. The following
6932 recording methods are available:
6936 Full record/replay recording using @value{GDBN}'s software record and
6937 replay implementation. This method allows replaying and reverse
6940 @item btrace @var{format}
6941 Hardware-supported instruction recording, supported on Intel
6942 processors. This method does not record data. Further, the data is
6943 collected in a ring buffer so old data will be overwritten when the
6944 buffer is full. It allows limited reverse execution. Variables and
6945 registers are not available during reverse execution. In remote
6946 debugging, recording continues on disconnect. Recorded data can be
6947 inspected after reconnecting. The recording may be stopped using
6950 The recording format can be specified as parameter. Without a parameter
6951 the command chooses the recording format. The following recording
6952 formats are available:
6956 @cindex branch trace store
6957 Use the @dfn{Branch Trace Store} (@acronym{BTS}) recording format. In
6958 this format, the processor stores a from/to record for each executed
6959 branch in the btrace ring buffer.
6962 @cindex Intel Processor Trace
6963 Use the @dfn{Intel Processor Trace} recording format. In this
6964 format, the processor stores the execution trace in a compressed form
6965 that is afterwards decoded by @value{GDBN}.
6967 The trace can be recorded with very low overhead. The compressed
6968 trace format also allows small trace buffers to already contain a big
6969 number of instructions compared to @acronym{BTS}.
6971 Decoding the recorded execution trace, on the other hand, is more
6972 expensive than decoding @acronym{BTS} trace. This is mostly due to the
6973 increased number of instructions to process. You should increase the
6974 buffer-size with care.
6977 Not all recording formats may be available on all processors.
6980 The process record and replay target can only debug a process that is
6981 already running. Therefore, you need first to start the process with
6982 the @kbd{run} or @kbd{start} commands, and then start the recording
6983 with the @kbd{record @var{method}} command.
6985 @cindex displaced stepping, and process record and replay
6986 Displaced stepping (@pxref{Maintenance Commands,, displaced stepping})
6987 will be automatically disabled when process record and replay target
6988 is started. That's because the process record and replay target
6989 doesn't support displaced stepping.
6991 @cindex non-stop mode, and process record and replay
6992 @cindex asynchronous execution, and process record and replay
6993 If the inferior is in the non-stop mode (@pxref{Non-Stop Mode}) or in
6994 the asynchronous execution mode (@pxref{Background Execution}), not
6995 all recording methods are available. The @code{full} recording method
6996 does not support these two modes.
7001 Stop the process record and replay target. When process record and
7002 replay target stops, the entire execution log will be deleted and the
7003 inferior will either be terminated, or will remain in its final state.
7005 When you stop the process record and replay target in record mode (at
7006 the end of the execution log), the inferior will be stopped at the
7007 next instruction that would have been recorded. In other words, if
7008 you record for a while and then stop recording, the inferior process
7009 will be left in the same state as if the recording never happened.
7011 On the other hand, if the process record and replay target is stopped
7012 while in replay mode (that is, not at the end of the execution log,
7013 but at some earlier point), the inferior process will become ``live''
7014 at that earlier state, and it will then be possible to continue the
7015 usual ``live'' debugging of the process from that state.
7017 When the inferior process exits, or @value{GDBN} detaches from it,
7018 process record and replay target will automatically stop itself.
7022 Go to a specific location in the execution log. There are several
7023 ways to specify the location to go to:
7026 @item record goto begin
7027 @itemx record goto start
7028 Go to the beginning of the execution log.
7030 @item record goto end
7031 Go to the end of the execution log.
7033 @item record goto @var{n}
7034 Go to instruction number @var{n} in the execution log.
7038 @item record save @var{filename}
7039 Save the execution log to a file @file{@var{filename}}.
7040 Default filename is @file{gdb_record.@var{process_id}}, where
7041 @var{process_id} is the process ID of the inferior.
7043 This command may not be available for all recording methods.
7045 @kindex record restore
7046 @item record restore @var{filename}
7047 Restore the execution log from a file @file{@var{filename}}.
7048 File must have been created with @code{record save}.
7050 @kindex set record full
7051 @item set record full insn-number-max @var{limit}
7052 @itemx set record full insn-number-max unlimited
7053 Set the limit of instructions to be recorded for the @code{full}
7054 recording method. Default value is 200000.
7056 If @var{limit} is a positive number, then @value{GDBN} will start
7057 deleting instructions from the log once the number of the record
7058 instructions becomes greater than @var{limit}. For every new recorded
7059 instruction, @value{GDBN} will delete the earliest recorded
7060 instruction to keep the number of recorded instructions at the limit.
7061 (Since deleting recorded instructions loses information, @value{GDBN}
7062 lets you control what happens when the limit is reached, by means of
7063 the @code{stop-at-limit} option, described below.)
7065 If @var{limit} is @code{unlimited} or zero, @value{GDBN} will never
7066 delete recorded instructions from the execution log. The number of
7067 recorded instructions is limited only by the available memory.
7069 @kindex show record full
7070 @item show record full insn-number-max
7071 Show the limit of instructions to be recorded with the @code{full}
7074 @item set record full stop-at-limit
7075 Control the behavior of the @code{full} recording method when the
7076 number of recorded instructions reaches the limit. If ON (the
7077 default), @value{GDBN} will stop when the limit is reached for the
7078 first time and ask you whether you want to stop the inferior or
7079 continue running it and recording the execution log. If you decide
7080 to continue recording, each new recorded instruction will cause the
7081 oldest one to be deleted.
7083 If this option is OFF, @value{GDBN} will automatically delete the
7084 oldest record to make room for each new one, without asking.
7086 @item show record full stop-at-limit
7087 Show the current setting of @code{stop-at-limit}.
7089 @item set record full memory-query
7090 Control the behavior when @value{GDBN} is unable to record memory
7091 changes caused by an instruction for the @code{full} recording method.
7092 If ON, @value{GDBN} will query whether to stop the inferior in that
7095 If this option is OFF (the default), @value{GDBN} will automatically
7096 ignore the effect of such instructions on memory. Later, when
7097 @value{GDBN} replays this execution log, it will mark the log of this
7098 instruction as not accessible, and it will not affect the replay
7101 @item show record full memory-query
7102 Show the current setting of @code{memory-query}.
7104 @kindex set record btrace
7105 The @code{btrace} record target does not trace data. As a
7106 convenience, when replaying, @value{GDBN} reads read-only memory off
7107 the live program directly, assuming that the addresses of the
7108 read-only areas don't change. This for example makes it possible to
7109 disassemble code while replaying, but not to print variables.
7110 In some cases, being able to inspect variables might be useful.
7111 You can use the following command for that:
7113 @item set record btrace replay-memory-access
7114 Control the behavior of the @code{btrace} recording method when
7115 accessing memory during replay. If @code{read-only} (the default),
7116 @value{GDBN} will only allow accesses to read-only memory.
7117 If @code{read-write}, @value{GDBN} will allow accesses to read-only
7118 and to read-write memory. Beware that the accessed memory corresponds
7119 to the live target and not necessarily to the current replay
7122 @item set record btrace cpu @var{identifier}
7123 Set the processor to be used for enabling workarounds for processor
7124 errata when decoding the trace.
7126 Processor errata are defects in processor operation, caused by its
7127 design or manufacture. They can cause a trace not to match the
7128 specification. This, in turn, may cause trace decode to fail.
7129 @value{GDBN} can detect erroneous trace packets and correct them, thus
7130 avoiding the decoding failures. These corrections are known as
7131 @dfn{errata workarounds}, and are enabled based on the processor on
7132 which the trace was recorded.
7134 By default, @value{GDBN} attempts to detect the processor
7135 automatically, and apply the necessary workarounds for it. However,
7136 you may need to specify the processor if @value{GDBN} does not yet
7137 support it. This command allows you to do that, and also allows to
7138 disable the workarounds.
7140 The argument @var{identifier} identifies the @sc{cpu} and is of the
7141 form: @code{@var{vendor}:@var{procesor identifier}}. In addition,
7142 there are two special identifiers, @code{none} and @code{auto}
7145 The following vendor identifiers and corresponding processor
7146 identifiers are currently supported:
7148 @multitable @columnfractions .1 .9
7151 @tab @var{family}/@var{model}[/@var{stepping}]
7155 On GNU/Linux systems, the processor @var{family}, @var{model}, and
7156 @var{stepping} can be obtained from @code{/proc/cpuinfo}.
7158 If @var{identifier} is @code{auto}, enable errata workarounds for the
7159 processor on which the trace was recorded. If @var{identifier} is
7160 @code{none}, errata workarounds are disabled.
7162 For example, when using an old @value{GDBN} on a new system, decode
7163 may fail because @value{GDBN} does not support the new processor. It
7164 often suffices to specify an older processor that @value{GDBN}
7169 Active record target: record-btrace
7170 Recording format: Intel Processor Trace.
7172 Failed to configure the Intel Processor Trace decoder: unknown cpu.
7173 (gdb) set record btrace cpu intel:6/158
7175 Active record target: record-btrace
7176 Recording format: Intel Processor Trace.
7178 Recorded 84872 instructions in 3189 functions (0 gaps) for thread 1 (...).
7181 @kindex show record btrace
7182 @item show record btrace replay-memory-access
7183 Show the current setting of @code{replay-memory-access}.
7185 @item show record btrace cpu
7186 Show the processor to be used for enabling trace decode errata
7189 @kindex set record btrace bts
7190 @item set record btrace bts buffer-size @var{size}
7191 @itemx set record btrace bts buffer-size unlimited
7192 Set the requested ring buffer size for branch tracing in @acronym{BTS}
7193 format. Default is 64KB.
7195 If @var{size} is a positive number, then @value{GDBN} will try to
7196 allocate a buffer of at least @var{size} bytes for each new thread
7197 that uses the btrace recording method and the @acronym{BTS} format.
7198 The actually obtained buffer size may differ from the requested
7199 @var{size}. Use the @code{info record} command to see the actual
7200 buffer size for each thread that uses the btrace recording method and
7201 the @acronym{BTS} format.
7203 If @var{limit} is @code{unlimited} or zero, @value{GDBN} will try to
7204 allocate a buffer of 4MB.
7206 Bigger buffers mean longer traces. On the other hand, @value{GDBN} will
7207 also need longer to process the branch trace data before it can be used.
7209 @item show record btrace bts buffer-size @var{size}
7210 Show the current setting of the requested ring buffer size for branch
7211 tracing in @acronym{BTS} format.
7213 @kindex set record btrace pt
7214 @item set record btrace pt buffer-size @var{size}
7215 @itemx set record btrace pt buffer-size unlimited
7216 Set the requested ring buffer size for branch tracing in Intel
7217 Processor Trace format. Default is 16KB.
7219 If @var{size} is a positive number, then @value{GDBN} will try to
7220 allocate a buffer of at least @var{size} bytes for each new thread
7221 that uses the btrace recording method and the Intel Processor Trace
7222 format. The actually obtained buffer size may differ from the
7223 requested @var{size}. Use the @code{info record} command to see the
7224 actual buffer size for each thread.
7226 If @var{limit} is @code{unlimited} or zero, @value{GDBN} will try to
7227 allocate a buffer of 4MB.
7229 Bigger buffers mean longer traces. On the other hand, @value{GDBN} will
7230 also need longer to process the branch trace data before it can be used.
7232 @item show record btrace pt buffer-size @var{size}
7233 Show the current setting of the requested ring buffer size for branch
7234 tracing in Intel Processor Trace format.
7238 Show various statistics about the recording depending on the recording
7243 For the @code{full} recording method, it shows the state of process
7244 record and its in-memory execution log buffer, including:
7248 Whether in record mode or replay mode.
7250 Lowest recorded instruction number (counting from when the current execution log started recording instructions).
7252 Highest recorded instruction number.
7254 Current instruction about to be replayed (if in replay mode).
7256 Number of instructions contained in the execution log.
7258 Maximum number of instructions that may be contained in the execution log.
7262 For the @code{btrace} recording method, it shows:
7268 Number of instructions that have been recorded.
7270 Number of blocks of sequential control-flow formed by the recorded
7273 Whether in record mode or replay mode.
7276 For the @code{bts} recording format, it also shows:
7279 Size of the perf ring buffer.
7282 For the @code{pt} recording format, it also shows:
7285 Size of the perf ring buffer.
7289 @kindex record delete
7292 When record target runs in replay mode (``in the past''), delete the
7293 subsequent execution log and begin to record a new execution log starting
7294 from the current address. This means you will abandon the previously
7295 recorded ``future'' and begin recording a new ``future''.
7297 @kindex record instruction-history
7298 @kindex rec instruction-history
7299 @item record instruction-history
7300 Disassembles instructions from the recorded execution log. By
7301 default, ten instructions are disassembled. This can be changed using
7302 the @code{set record instruction-history-size} command. Instructions
7303 are printed in execution order.
7305 It can also print mixed source+disassembly if you specify the the
7306 @code{/m} or @code{/s} modifier, and print the raw instructions in hex
7307 as well as in symbolic form by specifying the @code{/r} modifier.
7309 The current position marker is printed for the instruction at the
7310 current program counter value. This instruction can appear multiple
7311 times in the trace and the current position marker will be printed
7312 every time. To omit the current position marker, specify the
7315 To better align the printed instructions when the trace contains
7316 instructions from more than one function, the function name may be
7317 omitted by specifying the @code{/f} modifier.
7319 Speculatively executed instructions are prefixed with @samp{?}. This
7320 feature is not available for all recording formats.
7322 There are several ways to specify what part of the execution log to
7326 @item record instruction-history @var{insn}
7327 Disassembles ten instructions starting from instruction number
7330 @item record instruction-history @var{insn}, +/-@var{n}
7331 Disassembles @var{n} instructions around instruction number
7332 @var{insn}. If @var{n} is preceded with @code{+}, disassembles
7333 @var{n} instructions after instruction number @var{insn}. If
7334 @var{n} is preceded with @code{-}, disassembles @var{n}
7335 instructions before instruction number @var{insn}.
7337 @item record instruction-history
7338 Disassembles ten more instructions after the last disassembly.
7340 @item record instruction-history -
7341 Disassembles ten more instructions before the last disassembly.
7343 @item record instruction-history @var{begin}, @var{end}
7344 Disassembles instructions beginning with instruction number
7345 @var{begin} until instruction number @var{end}. The instruction
7346 number @var{end} is included.
7349 This command may not be available for all recording methods.
7352 @item set record instruction-history-size @var{size}
7353 @itemx set record instruction-history-size unlimited
7354 Define how many instructions to disassemble in the @code{record
7355 instruction-history} command. The default value is 10.
7356 A @var{size} of @code{unlimited} means unlimited instructions.
7359 @item show record instruction-history-size
7360 Show how many instructions to disassemble in the @code{record
7361 instruction-history} command.
7363 @kindex record function-call-history
7364 @kindex rec function-call-history
7365 @item record function-call-history
7366 Prints the execution history at function granularity. It prints one
7367 line for each sequence of instructions that belong to the same
7368 function giving the name of that function, the source lines
7369 for this instruction sequence (if the @code{/l} modifier is
7370 specified), and the instructions numbers that form the sequence (if
7371 the @code{/i} modifier is specified). The function names are indented
7372 to reflect the call stack depth if the @code{/c} modifier is
7373 specified. The @code{/l}, @code{/i}, and @code{/c} modifiers can be
7377 (@value{GDBP}) @b{list 1, 10}
7388 (@value{GDBP}) @b{record function-call-history /ilc}
7389 1 bar inst 1,4 at foo.c:6,8
7390 2 foo inst 5,10 at foo.c:2,3
7391 3 bar inst 11,13 at foo.c:9,10
7394 By default, ten lines are printed. This can be changed using the
7395 @code{set record function-call-history-size} command. Functions are
7396 printed in execution order. There are several ways to specify what
7400 @item record function-call-history @var{func}
7401 Prints ten functions starting from function number @var{func}.
7403 @item record function-call-history @var{func}, +/-@var{n}
7404 Prints @var{n} functions around function number @var{func}. If
7405 @var{n} is preceded with @code{+}, prints @var{n} functions after
7406 function number @var{func}. If @var{n} is preceded with @code{-},
7407 prints @var{n} functions before function number @var{func}.
7409 @item record function-call-history
7410 Prints ten more functions after the last ten-line print.
7412 @item record function-call-history -
7413 Prints ten more functions before the last ten-line print.
7415 @item record function-call-history @var{begin}, @var{end}
7416 Prints functions beginning with function number @var{begin} until
7417 function number @var{end}. The function number @var{end} is included.
7420 This command may not be available for all recording methods.
7422 @item set record function-call-history-size @var{size}
7423 @itemx set record function-call-history-size unlimited
7424 Define how many lines to print in the
7425 @code{record function-call-history} command. The default value is 10.
7426 A size of @code{unlimited} means unlimited lines.
7428 @item show record function-call-history-size
7429 Show how many lines to print in the
7430 @code{record function-call-history} command.
7435 @chapter Examining the Stack
7437 When your program has stopped, the first thing you need to know is where it
7438 stopped and how it got there.
7441 Each time your program performs a function call, information about the call
7443 That information includes the location of the call in your program,
7444 the arguments of the call,
7445 and the local variables of the function being called.
7446 The information is saved in a block of data called a @dfn{stack frame}.
7447 The stack frames are allocated in a region of memory called the @dfn{call
7450 When your program stops, the @value{GDBN} commands for examining the
7451 stack allow you to see all of this information.
7453 @cindex selected frame
7454 One of the stack frames is @dfn{selected} by @value{GDBN} and many
7455 @value{GDBN} commands refer implicitly to the selected frame. In
7456 particular, whenever you ask @value{GDBN} for the value of a variable in
7457 your program, the value is found in the selected frame. There are
7458 special @value{GDBN} commands to select whichever frame you are
7459 interested in. @xref{Selection, ,Selecting a Frame}.
7461 When your program stops, @value{GDBN} automatically selects the
7462 currently executing frame and describes it briefly, similar to the
7463 @code{frame} command (@pxref{Frame Info, ,Information about a Frame}).
7466 * Frames:: Stack frames
7467 * Backtrace:: Backtraces
7468 * Selection:: Selecting a frame
7469 * Frame Info:: Information on a frame
7470 * Frame Apply:: Applying a command to several frames
7471 * Frame Filter Management:: Managing frame filters
7476 @section Stack Frames
7478 @cindex frame, definition
7480 The call stack is divided up into contiguous pieces called @dfn{stack
7481 frames}, or @dfn{frames} for short; each frame is the data associated
7482 with one call to one function. The frame contains the arguments given
7483 to the function, the function's local variables, and the address at
7484 which the function is executing.
7486 @cindex initial frame
7487 @cindex outermost frame
7488 @cindex innermost frame
7489 When your program is started, the stack has only one frame, that of the
7490 function @code{main}. This is called the @dfn{initial} frame or the
7491 @dfn{outermost} frame. Each time a function is called, a new frame is
7492 made. Each time a function returns, the frame for that function invocation
7493 is eliminated. If a function is recursive, there can be many frames for
7494 the same function. The frame for the function in which execution is
7495 actually occurring is called the @dfn{innermost} frame. This is the most
7496 recently created of all the stack frames that still exist.
7498 @cindex frame pointer
7499 Inside your program, stack frames are identified by their addresses. A
7500 stack frame consists of many bytes, each of which has its own address; each
7501 kind of computer has a convention for choosing one byte whose
7502 address serves as the address of the frame. Usually this address is kept
7503 in a register called the @dfn{frame pointer register}
7504 (@pxref{Registers, $fp}) while execution is going on in that frame.
7507 @cindex frame number
7508 @value{GDBN} labels each existing stack frame with a @dfn{level}, a
7509 number that is zero for the innermost frame, one for the frame that
7510 called it, and so on upward. These level numbers give you a way of
7511 designating stack frames in @value{GDBN} commands. The terms
7512 @dfn{frame number} and @dfn{frame level} can be used interchangeably to
7513 describe this number.
7515 @c The -fomit-frame-pointer below perennially causes hbox overflow
7516 @c underflow problems.
7517 @cindex frameless execution
7518 Some compilers provide a way to compile functions so that they operate
7519 without stack frames. (For example, the @value{NGCC} option
7521 @samp{-fomit-frame-pointer}
7523 generates functions without a frame.)
7524 This is occasionally done with heavily used library functions to save
7525 the frame setup time. @value{GDBN} has limited facilities for dealing
7526 with these function invocations. If the innermost function invocation
7527 has no stack frame, @value{GDBN} nevertheless regards it as though
7528 it had a separate frame, which is numbered zero as usual, allowing
7529 correct tracing of the function call chain. However, @value{GDBN} has
7530 no provision for frameless functions elsewhere in the stack.
7536 @cindex call stack traces
7537 A backtrace is a summary of how your program got where it is. It shows one
7538 line per frame, for many frames, starting with the currently executing
7539 frame (frame zero), followed by its caller (frame one), and on up the
7542 @anchor{backtrace-command}
7544 @kindex bt @r{(@code{backtrace})}
7545 To print a backtrace of the entire stack, use the @code{backtrace}
7546 command, or its alias @code{bt}. This command will print one line per
7547 frame for frames in the stack. By default, all stack frames are
7548 printed. You can stop the backtrace at any time by typing the system
7549 interrupt character, normally @kbd{Ctrl-c}.
7552 @item backtrace [@var{args}@dots{}]
7553 @itemx bt [@var{args}@dots{}]
7554 Print the backtrace of the entire stack. The optional @var{args} can
7555 be one of the following:
7560 Print only the innermost @var{n} frames, where @var{n} is a positive
7565 Print only the outermost @var{n} frames, where @var{n} is a positive
7569 Print the values of the local variables also. This can be combined
7570 with a number to limit the number of frames shown.
7573 Do not run Python frame filters on this backtrace. @xref{Frame
7574 Filter API}, for more information. Additionally use @ref{disable
7575 frame-filter all} to turn off all frame filters. This is only
7576 relevant when @value{GDBN} has been configured with @code{Python}
7580 A Python frame filter might decide to ``elide'' some frames. Normally
7581 such elided frames are still printed, but they are indented relative
7582 to the filtered frames that cause them to be elided. The @code{hide}
7583 option causes elided frames to not be printed at all.
7589 The names @code{where} and @code{info stack} (abbreviated @code{info s})
7590 are additional aliases for @code{backtrace}.
7592 @cindex multiple threads, backtrace
7593 In a multi-threaded program, @value{GDBN} by default shows the
7594 backtrace only for the current thread. To display the backtrace for
7595 several or all of the threads, use the command @code{thread apply}
7596 (@pxref{Threads, thread apply}). For example, if you type @kbd{thread
7597 apply all backtrace}, @value{GDBN} will display the backtrace for all
7598 the threads; this is handy when you debug a core dump of a
7599 multi-threaded program.
7601 Each line in the backtrace shows the frame number and the function name.
7602 The program counter value is also shown---unless you use @code{set
7603 print address off}. The backtrace also shows the source file name and
7604 line number, as well as the arguments to the function. The program
7605 counter value is omitted if it is at the beginning of the code for that
7608 Here is an example of a backtrace. It was made with the command
7609 @samp{bt 3}, so it shows the innermost three frames.
7613 #0 m4_traceon (obs=0x24eb0, argc=1, argv=0x2b8c8)
7615 #1 0x6e38 in expand_macro (sym=0x2b600, data=...) at macro.c:242
7616 #2 0x6840 in expand_token (obs=0x0, t=177664, td=0xf7fffb08)
7618 (More stack frames follow...)
7623 The display for frame zero does not begin with a program counter
7624 value, indicating that your program has stopped at the beginning of the
7625 code for line @code{993} of @code{builtin.c}.
7628 The value of parameter @code{data} in frame 1 has been replaced by
7629 @code{@dots{}}. By default, @value{GDBN} prints the value of a parameter
7630 only if it is a scalar (integer, pointer, enumeration, etc). See command
7631 @kbd{set print frame-arguments} in @ref{Print Settings} for more details
7632 on how to configure the way function parameter values are printed.
7634 @cindex optimized out, in backtrace
7635 @cindex function call arguments, optimized out
7636 If your program was compiled with optimizations, some compilers will
7637 optimize away arguments passed to functions if those arguments are
7638 never used after the call. Such optimizations generate code that
7639 passes arguments through registers, but doesn't store those arguments
7640 in the stack frame. @value{GDBN} has no way of displaying such
7641 arguments in stack frames other than the innermost one. Here's what
7642 such a backtrace might look like:
7646 #0 m4_traceon (obs=0x24eb0, argc=1, argv=0x2b8c8)
7648 #1 0x6e38 in expand_macro (sym=<optimized out>) at macro.c:242
7649 #2 0x6840 in expand_token (obs=0x0, t=<optimized out>, td=0xf7fffb08)
7651 (More stack frames follow...)
7656 The values of arguments that were not saved in their stack frames are
7657 shown as @samp{<optimized out>}.
7659 If you need to display the values of such optimized-out arguments,
7660 either deduce that from other variables whose values depend on the one
7661 you are interested in, or recompile without optimizations.
7663 @cindex backtrace beyond @code{main} function
7664 @cindex program entry point
7665 @cindex startup code, and backtrace
7666 Most programs have a standard user entry point---a place where system
7667 libraries and startup code transition into user code. For C this is
7668 @code{main}@footnote{
7669 Note that embedded programs (the so-called ``free-standing''
7670 environment) are not required to have a @code{main} function as the
7671 entry point. They could even have multiple entry points.}.
7672 When @value{GDBN} finds the entry function in a backtrace
7673 it will terminate the backtrace, to avoid tracing into highly
7674 system-specific (and generally uninteresting) code.
7676 If you need to examine the startup code, or limit the number of levels
7677 in a backtrace, you can change this behavior:
7680 @item set backtrace past-main
7681 @itemx set backtrace past-main on
7682 @kindex set backtrace
7683 Backtraces will continue past the user entry point.
7685 @item set backtrace past-main off
7686 Backtraces will stop when they encounter the user entry point. This is the
7689 @item show backtrace past-main
7690 @kindex show backtrace
7691 Display the current user entry point backtrace policy.
7693 @item set backtrace past-entry
7694 @itemx set backtrace past-entry on
7695 Backtraces will continue past the internal entry point of an application.
7696 This entry point is encoded by the linker when the application is built,
7697 and is likely before the user entry point @code{main} (or equivalent) is called.
7699 @item set backtrace past-entry off
7700 Backtraces will stop when they encounter the internal entry point of an
7701 application. This is the default.
7703 @item show backtrace past-entry
7704 Display the current internal entry point backtrace policy.
7706 @item set backtrace limit @var{n}
7707 @itemx set backtrace limit 0
7708 @itemx set backtrace limit unlimited
7709 @cindex backtrace limit
7710 Limit the backtrace to @var{n} levels. A value of @code{unlimited}
7711 or zero means unlimited levels.
7713 @item show backtrace limit
7714 Display the current limit on backtrace levels.
7717 You can control how file names are displayed.
7720 @item set filename-display
7721 @itemx set filename-display relative
7722 @cindex filename-display
7723 Display file names relative to the compilation directory. This is the default.
7725 @item set filename-display basename
7726 Display only basename of a filename.
7728 @item set filename-display absolute
7729 Display an absolute filename.
7731 @item show filename-display
7732 Show the current way to display filenames.
7736 @section Selecting a Frame
7738 Most commands for examining the stack and other data in your program work on
7739 whichever stack frame is selected at the moment. Here are the commands for
7740 selecting a stack frame; all of them finish by printing a brief description
7741 of the stack frame just selected.
7744 @kindex frame@r{, selecting}
7745 @kindex f @r{(@code{frame})}
7746 @item frame @r{[} @var{frame-selection-spec} @r{]}
7747 @item f @r{[} @var{frame-selection-spec} @r{]}
7748 The @command{frame} command allows different stack frames to be
7749 selected. The @var{frame-selection-spec} can be any of the following:
7754 @item level @var{num}
7755 Select frame level @var{num}. Recall that frame zero is the innermost
7756 (currently executing) frame, frame one is the frame that called the
7757 innermost one, and so on. The highest level frame is usually the one
7760 As this is the most common method of navigating the frame stack, the
7761 string @command{level} can be omitted. For example, the following two
7762 commands are equivalent:
7765 (@value{GDBP}) frame 3
7766 (@value{GDBP}) frame level 3
7769 @kindex frame address
7770 @item address @var{stack-address}
7771 Select the frame with stack address @var{stack-address}. The
7772 @var{stack-address} for a frame can be seen in the output of
7773 @command{info frame}, for example:
7777 Stack level 1, frame at 0x7fffffffda30:
7778 rip = 0x40066d in b (amd64-entry-value.cc:59); saved rip 0x4004c5
7779 tail call frame, caller of frame at 0x7fffffffda30
7780 source language c++.
7781 Arglist at unknown address.
7782 Locals at unknown address, Previous frame's sp is 0x7fffffffda30
7785 The @var{stack-address} for this frame is @code{0x7fffffffda30} as
7786 indicated by the line:
7789 Stack level 1, frame at 0x7fffffffda30:
7792 @kindex frame function
7793 @item function @var{function-name}
7794 Select the stack frame for function @var{function-name}. If there are
7795 multiple stack frames for function @var{function-name} then the inner
7796 most stack frame is selected.
7799 @item view @var{stack-address} @r{[} @var{pc-addr} @r{]}
7800 View a frame that is not part of @value{GDBN}'s backtrace. The frame
7801 viewed has stack address @var{stack-addr}, and optionally, a program
7802 counter address of @var{pc-addr}.
7804 This is useful mainly if the chaining of stack frames has been
7805 damaged by a bug, making it impossible for @value{GDBN} to assign
7806 numbers properly to all frames. In addition, this can be useful
7807 when your program has multiple stacks and switches between them.
7809 When viewing a frame outside the current backtrace using
7810 @command{frame view} then you can always return to the original
7811 stack using one of the previous stack frame selection instructions,
7812 for example @command{frame level 0}.
7818 Move @var{n} frames up the stack; @var{n} defaults to 1. For positive
7819 numbers @var{n}, this advances toward the outermost frame, to higher
7820 frame numbers, to frames that have existed longer.
7823 @kindex do @r{(@code{down})}
7825 Move @var{n} frames down the stack; @var{n} defaults to 1. For
7826 positive numbers @var{n}, this advances toward the innermost frame, to
7827 lower frame numbers, to frames that were created more recently.
7828 You may abbreviate @code{down} as @code{do}.
7831 All of these commands end by printing two lines of output describing the
7832 frame. The first line shows the frame number, the function name, the
7833 arguments, and the source file and line number of execution in that
7834 frame. The second line shows the text of that source line.
7842 #1 0x22f0 in main (argc=1, argv=0xf7fffbf4, env=0xf7fffbfc)
7844 10 read_input_file (argv[i]);
7848 After such a printout, the @code{list} command with no arguments
7849 prints ten lines centered on the point of execution in the frame.
7850 You can also edit the program at the point of execution with your favorite
7851 editing program by typing @code{edit}.
7852 @xref{List, ,Printing Source Lines},
7856 @kindex select-frame
7857 @item select-frame @r{[} @var{frame-selection-spec} @r{]}
7858 The @code{select-frame} command is a variant of @code{frame} that does
7859 not display the new frame after selecting it. This command is
7860 intended primarily for use in @value{GDBN} command scripts, where the
7861 output might be unnecessary and distracting. The
7862 @var{frame-selection-spec} is as for the @command{frame} command
7863 described in @ref{Selection, ,Selecting a Frame}.
7865 @kindex down-silently
7867 @item up-silently @var{n}
7868 @itemx down-silently @var{n}
7869 These two commands are variants of @code{up} and @code{down},
7870 respectively; they differ in that they do their work silently, without
7871 causing display of the new frame. They are intended primarily for use
7872 in @value{GDBN} command scripts, where the output might be unnecessary and
7877 @section Information About a Frame
7879 There are several other commands to print information about the selected
7885 When used without any argument, this command does not change which
7886 frame is selected, but prints a brief description of the currently
7887 selected stack frame. It can be abbreviated @code{f}. With an
7888 argument, this command is used to select a stack frame.
7889 @xref{Selection, ,Selecting a Frame}.
7892 @kindex info f @r{(@code{info frame})}
7895 This command prints a verbose description of the selected stack frame,
7900 the address of the frame
7902 the address of the next frame down (called by this frame)
7904 the address of the next frame up (caller of this frame)
7906 the language in which the source code corresponding to this frame is written
7908 the address of the frame's arguments
7910 the address of the frame's local variables
7912 the program counter saved in it (the address of execution in the caller frame)
7914 which registers were saved in the frame
7917 @noindent The verbose description is useful when
7918 something has gone wrong that has made the stack format fail to fit
7919 the usual conventions.
7921 @item info frame @r{[} @var{frame-selection-spec} @r{]}
7922 @itemx info f @r{[} @var{frame-selection-spec} @r{]}
7923 Print a verbose description of the frame selected by
7924 @var{frame-selection-spec}. The @var{frame-selection-spec} is the
7925 same as for the @command{frame} command (@pxref{Selection, ,Selecting
7926 a Frame}). The selected frame remains unchanged by this command.
7929 @item info args [-q]
7930 Print the arguments of the selected frame, each on a separate line.
7932 The optional flag @samp{-q}, which stands for @samp{quiet}, disables
7933 printing header information and messages explaining why no argument
7936 @item info args [-q] [-t @var{type_regexp}] [@var{regexp}]
7937 Like @kbd{info args}, but only print the arguments selected
7938 with the provided regexp(s).
7940 If @var{regexp} is provided, print only the arguments whose names
7941 match the regular expression @var{regexp}.
7943 If @var{type_regexp} is provided, print only the arguments whose
7944 types, as printed by the @code{whatis} command, match
7945 the regular expression @var{type_regexp}.
7946 If @var{type_regexp} contains space(s), it should be enclosed in
7947 quote characters. If needed, use backslash to escape the meaning
7948 of special characters or quotes.
7950 If both @var{regexp} and @var{type_regexp} are provided, an argument
7951 is printed only if its name matches @var{regexp} and its type matches
7954 @item info locals [-q]
7956 Print the local variables of the selected frame, each on a separate
7957 line. These are all variables (declared either static or automatic)
7958 accessible at the point of execution of the selected frame.
7960 The optional flag @samp{-q}, which stands for @samp{quiet}, disables
7961 printing header information and messages explaining why no local variables
7964 @item info locals [-q] [-t @var{type_regexp}] [@var{regexp}]
7965 Like @kbd{info locals}, but only print the local variables selected
7966 with the provided regexp(s).
7968 If @var{regexp} is provided, print only the local variables whose names
7969 match the regular expression @var{regexp}.
7971 If @var{type_regexp} is provided, print only the local variables 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, a local variable
7979 is printed only if its name matches @var{regexp} and its type matches
7982 The command @kbd{info locals -q -t @var{type_regexp}} can usefully be
7983 combined with the commands @kbd{frame apply} and @kbd{thread apply}.
7984 For example, your program might use Resource Acquisition Is
7985 Initialization types (RAII) such as @code{lock_something_t}: each
7986 local variable of type @code{lock_something_t} automatically places a
7987 lock that is destroyed when the variable goes out of scope. You can
7988 then list all acquired locks in your program by doing
7990 thread apply all -s frame apply all -s info locals -q -t lock_something_t
7993 or the equivalent shorter form
7995 tfaas i lo -q -t lock_something_t
8001 @section Applying a Command to Several Frames.
8003 @cindex apply command to several frames
8005 @item frame apply [all | @var{count} | @var{-count} | level @var{level}@dots{}] [@var{flag}]@dots{} @var{command}
8006 The @code{frame apply} command allows you to apply the named
8007 @var{command} to one or more frames.
8011 Specify @code{all} to apply @var{command} to all frames.
8014 Use @var{count} to apply @var{command} to the innermost @var{count}
8015 frames, where @var{count} is a positive number.
8018 Use @var{-count} to apply @var{command} to the outermost @var{count}
8019 frames, where @var{count} is a positive number.
8022 Use @code{level} to apply @var{command} to the set of frames identified
8023 by the @var{level} list. @var{level} is a frame level or a range of frame
8024 levels as @var{level1}-@var{level2}. The frame level is the number shown
8025 in the first field of the @samp{backtrace} command output.
8026 E.g., @samp{2-4 6-8 3} indicates to apply @var{command} for the frames
8027 at levels 2, 3, 4, 6, 7, 8, and then again on frame at level 3.
8033 Note that the frames on which @code{frame apply} applies a command are
8034 also influenced by the @code{set backtrace} settings such as @code{set
8035 backtrace past-main} and @code{set backtrace limit N}. See
8036 @xref{Backtrace,,Backtraces}.
8038 The @var{flag} arguments control what output to produce and how to handle
8039 errors raised when applying @var{command} to a frame. @var{flag}
8040 must start with a @code{-} directly followed by one letter in
8041 @code{qcs}. If several flags are provided, they must be given
8042 individually, such as @code{-c -q}.
8044 By default, @value{GDBN} displays some frame information before the
8045 output produced by @var{command}, and an error raised during the
8046 execution of a @var{command} will abort @code{frame apply}. The
8047 following flags can be used to fine-tune this behavior:
8051 The flag @code{-c}, which stands for @samp{continue}, causes any
8052 errors in @var{command} to be displayed, and the execution of
8053 @code{frame apply} then continues.
8055 The flag @code{-s}, which stands for @samp{silent}, causes any errors
8056 or empty output produced by a @var{command} to be silently ignored.
8057 That is, the execution continues, but the frame information and errors
8060 The flag @code{-q} (@samp{quiet}) disables printing the frame
8064 The following example shows how the flags @code{-c} and @code{-s} are
8065 working when applying the command @code{p j} to all frames, where
8066 variable @code{j} can only be successfully printed in the outermost
8067 @code{#1 main} frame.
8071 (gdb) frame apply all p j
8072 #0 some_function (i=5) at fun.c:4
8073 No symbol "j" in current context.
8074 (gdb) frame apply all -c p j
8075 #0 some_function (i=5) at fun.c:4
8076 No symbol "j" in current context.
8077 #1 0x565555fb in main (argc=1, argv=0xffffd2c4) at fun.c:11
8079 (gdb) frame apply all -s p j
8080 #1 0x565555fb in main (argc=1, argv=0xffffd2c4) at fun.c:11
8086 By default, @samp{frame apply}, prints the frame location
8087 information before the command output:
8091 (gdb) frame apply all p $sp
8092 #0 some_function (i=5) at fun.c:4
8093 $4 = (void *) 0xffffd1e0
8094 #1 0x565555fb in main (argc=1, argv=0xffffd2c4) at fun.c:11
8095 $5 = (void *) 0xffffd1f0
8100 If flag @code{-q} is given, no frame information is printed:
8103 (gdb) frame apply all -q p $sp
8104 $12 = (void *) 0xffffd1e0
8105 $13 = (void *) 0xffffd1f0
8113 @cindex apply a command to all frames (ignoring errors and empty output)
8114 @item faas @var{command}
8115 Shortcut for @code{frame apply all -s @var{command}}.
8116 Applies @var{command} on all frames, ignoring errors and empty output.
8118 It can for example be used to print a local variable or a function
8119 argument without knowing the frame where this variable or argument
8122 (@value{GDBP}) faas p some_local_var_i_do_not_remember_where_it_is
8125 Note that the command @code{tfaas @var{command}} applies @var{command}
8126 on all frames of all threads. See @xref{Threads,,Threads}.
8130 @node Frame Filter Management
8131 @section Management of Frame Filters.
8132 @cindex managing frame filters
8134 Frame filters are Python based utilities to manage and decorate the
8135 output of frames. @xref{Frame Filter API}, for further information.
8137 Managing frame filters is performed by several commands available
8138 within @value{GDBN}, detailed here.
8141 @kindex info frame-filter
8142 @item info frame-filter
8143 Print a list of installed frame filters from all dictionaries, showing
8144 their name, priority and enabled status.
8146 @kindex disable frame-filter
8147 @anchor{disable frame-filter all}
8148 @item disable frame-filter @var{filter-dictionary} @var{filter-name}
8149 Disable a frame filter in the dictionary matching
8150 @var{filter-dictionary} and @var{filter-name}. The
8151 @var{filter-dictionary} may be @code{all}, @code{global},
8152 @code{progspace}, or the name of the object file where the frame filter
8153 dictionary resides. When @code{all} is specified, all frame filters
8154 across all dictionaries are disabled. The @var{filter-name} is the name
8155 of the frame filter and is used when @code{all} is not the option for
8156 @var{filter-dictionary}. A disabled frame-filter is not deleted, it
8157 may be enabled again later.
8159 @kindex enable frame-filter
8160 @item enable frame-filter @var{filter-dictionary} @var{filter-name}
8161 Enable a frame filter in the dictionary matching
8162 @var{filter-dictionary} and @var{filter-name}. The
8163 @var{filter-dictionary} may be @code{all}, @code{global},
8164 @code{progspace} or the name of the object file where the frame filter
8165 dictionary resides. When @code{all} is specified, all frame filters across
8166 all dictionaries are enabled. The @var{filter-name} is the name of the frame
8167 filter and is used when @code{all} is not the option for
8168 @var{filter-dictionary}.
8173 (gdb) info frame-filter
8175 global frame-filters:
8176 Priority Enabled Name
8177 1000 No PrimaryFunctionFilter
8180 progspace /build/test frame-filters:
8181 Priority Enabled Name
8182 100 Yes ProgspaceFilter
8184 objfile /build/test frame-filters:
8185 Priority Enabled Name
8186 999 Yes BuildProgra Filter
8188 (gdb) disable frame-filter /build/test BuildProgramFilter
8189 (gdb) info frame-filter
8191 global frame-filters:
8192 Priority Enabled Name
8193 1000 No PrimaryFunctionFilter
8196 progspace /build/test frame-filters:
8197 Priority Enabled Name
8198 100 Yes ProgspaceFilter
8200 objfile /build/test frame-filters:
8201 Priority Enabled Name
8202 999 No BuildProgramFilter
8204 (gdb) enable frame-filter global PrimaryFunctionFilter
8205 (gdb) info frame-filter
8207 global frame-filters:
8208 Priority Enabled Name
8209 1000 Yes PrimaryFunctionFilter
8212 progspace /build/test frame-filters:
8213 Priority Enabled Name
8214 100 Yes ProgspaceFilter
8216 objfile /build/test frame-filters:
8217 Priority Enabled Name
8218 999 No BuildProgramFilter
8221 @kindex set frame-filter priority
8222 @item set frame-filter priority @var{filter-dictionary} @var{filter-name} @var{priority}
8223 Set the @var{priority} of a frame filter in the dictionary matching
8224 @var{filter-dictionary}, and the frame filter name matching
8225 @var{filter-name}. The @var{filter-dictionary} may be @code{global},
8226 @code{progspace} or the name of the object file where the frame filter
8227 dictionary resides. The @var{priority} is an integer.
8229 @kindex show frame-filter priority
8230 @item show frame-filter priority @var{filter-dictionary} @var{filter-name}
8231 Show the @var{priority} of a frame filter in the dictionary matching
8232 @var{filter-dictionary}, and the frame filter name matching
8233 @var{filter-name}. The @var{filter-dictionary} may be @code{global},
8234 @code{progspace} or the name of the object file where the frame filter
8240 (gdb) info frame-filter
8242 global frame-filters:
8243 Priority Enabled Name
8244 1000 Yes PrimaryFunctionFilter
8247 progspace /build/test frame-filters:
8248 Priority Enabled Name
8249 100 Yes ProgspaceFilter
8251 objfile /build/test frame-filters:
8252 Priority Enabled Name
8253 999 No BuildProgramFilter
8255 (gdb) set frame-filter priority global Reverse 50
8256 (gdb) info frame-filter
8258 global frame-filters:
8259 Priority Enabled Name
8260 1000 Yes PrimaryFunctionFilter
8263 progspace /build/test frame-filters:
8264 Priority Enabled Name
8265 100 Yes ProgspaceFilter
8267 objfile /build/test frame-filters:
8268 Priority Enabled Name
8269 999 No BuildProgramFilter
8274 @chapter Examining Source Files
8276 @value{GDBN} can print parts of your program's source, since the debugging
8277 information recorded in the program tells @value{GDBN} what source files were
8278 used to build it. When your program stops, @value{GDBN} spontaneously prints
8279 the line where it stopped. Likewise, when you select a stack frame
8280 (@pxref{Selection, ,Selecting a Frame}), @value{GDBN} prints the line where
8281 execution in that frame has stopped. You can print other portions of
8282 source files by explicit command.
8284 If you use @value{GDBN} through its @sc{gnu} Emacs interface, you may
8285 prefer to use Emacs facilities to view source; see @ref{Emacs, ,Using
8286 @value{GDBN} under @sc{gnu} Emacs}.
8289 * List:: Printing source lines
8290 * Specify Location:: How to specify code locations
8291 * Edit:: Editing source files
8292 * Search:: Searching source files
8293 * Source Path:: Specifying source directories
8294 * Machine Code:: Source and machine code
8298 @section Printing Source Lines
8301 @kindex l @r{(@code{list})}
8302 To print lines from a source file, use the @code{list} command
8303 (abbreviated @code{l}). By default, ten lines are printed.
8304 There are several ways to specify what part of the file you want to
8305 print; see @ref{Specify Location}, for the full list.
8307 Here are the forms of the @code{list} command most commonly used:
8310 @item list @var{linenum}
8311 Print lines centered around line number @var{linenum} in the
8312 current source file.
8314 @item list @var{function}
8315 Print lines centered around the beginning of function
8319 Print more lines. If the last lines printed were printed with a
8320 @code{list} command, this prints lines following the last lines
8321 printed; however, if the last line printed was a solitary line printed
8322 as part of displaying a stack frame (@pxref{Stack, ,Examining the
8323 Stack}), this prints lines centered around that line.
8326 Print lines just before the lines last printed.
8329 @cindex @code{list}, how many lines to display
8330 By default, @value{GDBN} prints ten source lines with any of these forms of
8331 the @code{list} command. You can change this using @code{set listsize}:
8334 @kindex set listsize
8335 @item set listsize @var{count}
8336 @itemx set listsize unlimited
8337 Make the @code{list} command display @var{count} source lines (unless
8338 the @code{list} argument explicitly specifies some other number).
8339 Setting @var{count} to @code{unlimited} or 0 means there's no limit.
8341 @kindex show listsize
8343 Display the number of lines that @code{list} prints.
8346 Repeating a @code{list} command with @key{RET} discards the argument,
8347 so it is equivalent to typing just @code{list}. This is more useful
8348 than listing the same lines again. An exception is made for an
8349 argument of @samp{-}; that argument is preserved in repetition so that
8350 each repetition moves up in the source file.
8352 In general, the @code{list} command expects you to supply zero, one or two
8353 @dfn{locations}. Locations specify source lines; there are several ways
8354 of writing them (@pxref{Specify Location}), but the effect is always
8355 to specify some source line.
8357 Here is a complete description of the possible arguments for @code{list}:
8360 @item list @var{location}
8361 Print lines centered around the line specified by @var{location}.
8363 @item list @var{first},@var{last}
8364 Print lines from @var{first} to @var{last}. Both arguments are
8365 locations. When a @code{list} command has two locations, and the
8366 source file of the second location is omitted, this refers to
8367 the same source file as the first location.
8369 @item list ,@var{last}
8370 Print lines ending with @var{last}.
8372 @item list @var{first},
8373 Print lines starting with @var{first}.
8376 Print lines just after the lines last printed.
8379 Print lines just before the lines last printed.
8382 As described in the preceding table.
8385 @node Specify Location
8386 @section Specifying a Location
8387 @cindex specifying location
8389 @cindex source location
8392 * Linespec Locations:: Linespec locations
8393 * Explicit Locations:: Explicit locations
8394 * Address Locations:: Address locations
8397 Several @value{GDBN} commands accept arguments that specify a location
8398 of your program's code. Since @value{GDBN} is a source-level
8399 debugger, a location usually specifies some line in the source code.
8400 Locations may be specified using three different formats:
8401 linespec locations, explicit locations, or address locations.
8403 @node Linespec Locations
8404 @subsection Linespec Locations
8405 @cindex linespec locations
8407 A @dfn{linespec} is a colon-separated list of source location parameters such
8408 as file name, function name, etc. Here are all the different ways of
8409 specifying a linespec:
8413 Specifies the line number @var{linenum} of the current source file.
8416 @itemx +@var{offset}
8417 Specifies the line @var{offset} lines before or after the @dfn{current
8418 line}. For the @code{list} command, the current line is the last one
8419 printed; for the breakpoint commands, this is the line at which
8420 execution stopped in the currently selected @dfn{stack frame}
8421 (@pxref{Frames, ,Frames}, for a description of stack frames.) When
8422 used as the second of the two linespecs in a @code{list} command,
8423 this specifies the line @var{offset} lines up or down from the first
8426 @item @var{filename}:@var{linenum}
8427 Specifies the line @var{linenum} in the source file @var{filename}.
8428 If @var{filename} is a relative file name, then it will match any
8429 source file name with the same trailing components. For example, if
8430 @var{filename} is @samp{gcc/expr.c}, then it will match source file
8431 name of @file{/build/trunk/gcc/expr.c}, but not
8432 @file{/build/trunk/libcpp/expr.c} or @file{/build/trunk/gcc/x-expr.c}.
8434 @item @var{function}
8435 Specifies the line that begins the body of the function @var{function}.
8436 For example, in C, this is the line with the open brace.
8438 By default, in C@t{++} and Ada, @var{function} is interpreted as
8439 specifying all functions named @var{function} in all scopes. For
8440 C@t{++}, this means in all namespaces and classes. For Ada, this
8441 means in all packages.
8443 For example, assuming a program with C@t{++} symbols named
8444 @code{A::B::func} and @code{B::func}, both commands @w{@kbd{break
8445 func}} and @w{@kbd{break B::func}} set a breakpoint on both symbols.
8447 Commands that accept a linespec let you override this with the
8448 @code{-qualified} option. For example, @w{@kbd{break -qualified
8449 func}} sets a breakpoint on a free-function named @code{func} ignoring
8450 any C@t{++} class methods and namespace functions called @code{func}.
8452 @xref{Explicit Locations}.
8454 @item @var{function}:@var{label}
8455 Specifies the line where @var{label} appears in @var{function}.
8457 @item @var{filename}:@var{function}
8458 Specifies the line that begins the body of the function @var{function}
8459 in the file @var{filename}. You only need the file name with a
8460 function name to avoid ambiguity when there are identically named
8461 functions in different source files.
8464 Specifies the line at which the label named @var{label} appears
8465 in the function corresponding to the currently selected stack frame.
8466 If there is no current selected stack frame (for instance, if the inferior
8467 is not running), then @value{GDBN} will not search for a label.
8469 @cindex breakpoint at static probe point
8470 @item -pstap|-probe-stap @r{[}@var{objfile}:@r{[}@var{provider}:@r{]}@r{]}@var{name}
8471 The @sc{gnu}/Linux tool @code{SystemTap} provides a way for
8472 applications to embed static probes. @xref{Static Probe Points}, for more
8473 information on finding and using static probes. This form of linespec
8474 specifies the location of such a static probe.
8476 If @var{objfile} is given, only probes coming from that shared library
8477 or executable matching @var{objfile} as a regular expression are considered.
8478 If @var{provider} is given, then only probes from that provider are considered.
8479 If several probes match the spec, @value{GDBN} will insert a breakpoint at
8480 each one of those probes.
8483 @node Explicit Locations
8484 @subsection Explicit Locations
8485 @cindex explicit locations
8487 @dfn{Explicit locations} allow the user to directly specify the source
8488 location's parameters using option-value pairs.
8490 Explicit locations are useful when several functions, labels, or
8491 file names have the same name (base name for files) in the program's
8492 sources. In these cases, explicit locations point to the source
8493 line you meant more accurately and unambiguously. Also, using
8494 explicit locations might be faster in large programs.
8496 For example, the linespec @samp{foo:bar} may refer to a function @code{bar}
8497 defined in the file named @file{foo} or the label @code{bar} in a function
8498 named @code{foo}. @value{GDBN} must search either the file system or
8499 the symbol table to know.
8501 The list of valid explicit location options is summarized in the
8505 @item -source @var{filename}
8506 The value specifies the source file name. To differentiate between
8507 files with the same base name, prepend as many directories as is necessary
8508 to uniquely identify the desired file, e.g., @file{foo/bar/baz.c}. Otherwise
8509 @value{GDBN} will use the first file it finds with the given base
8510 name. This option requires the use of either @code{-function} or @code{-line}.
8512 @item -function @var{function}
8513 The value specifies the name of a function. Operations
8514 on function locations unmodified by other options (such as @code{-label}
8515 or @code{-line}) refer to the line that begins the body of the function.
8516 In C, for example, this is the line with the open brace.
8518 By default, in C@t{++} and Ada, @var{function} is interpreted as
8519 specifying all functions named @var{function} in all scopes. For
8520 C@t{++}, this means in all namespaces and classes. For Ada, this
8521 means in all packages.
8523 For example, assuming a program with C@t{++} symbols named
8524 @code{A::B::func} and @code{B::func}, both commands @w{@kbd{break
8525 -function func}} and @w{@kbd{break -function B::func}} set a
8526 breakpoint on both symbols.
8528 You can use the @kbd{-qualified} flag to override this (see below).
8532 This flag makes @value{GDBN} interpret a function name specified with
8533 @kbd{-function} as a complete fully-qualified name.
8535 For example, assuming a C@t{++} program with symbols named
8536 @code{A::B::func} and @code{B::func}, the @w{@kbd{break -qualified
8537 -function B::func}} command sets a breakpoint on @code{B::func}, only.
8539 (Note: the @kbd{-qualified} option can precede a linespec as well
8540 (@pxref{Linespec Locations}), so the particular example above could be
8541 simplified as @w{@kbd{break -qualified B::func}}.)
8543 @item -label @var{label}
8544 The value specifies the name of a label. When the function
8545 name is not specified, the label is searched in the function of the currently
8546 selected stack frame.
8548 @item -line @var{number}
8549 The value specifies a line offset for the location. The offset may either
8550 be absolute (@code{-line 3}) or relative (@code{-line +3}), depending on
8551 the command. When specified without any other options, the line offset is
8552 relative to the current line.
8555 Explicit location options may be abbreviated by omitting any non-unique
8556 trailing characters from the option name, e.g., @w{@kbd{break -s main.c -li 3}}.
8558 @node Address Locations
8559 @subsection Address Locations
8560 @cindex address locations
8562 @dfn{Address locations} indicate a specific program address. They have
8563 the generalized form *@var{address}.
8565 For line-oriented commands, such as @code{list} and @code{edit}, this
8566 specifies a source line that contains @var{address}. For @code{break} and
8567 other breakpoint-oriented commands, this can be used to set breakpoints in
8568 parts of your program which do not have debugging information or
8571 Here @var{address} may be any expression valid in the current working
8572 language (@pxref{Languages, working language}) that specifies a code
8573 address. In addition, as a convenience, @value{GDBN} extends the
8574 semantics of expressions used in locations to cover several situations
8575 that frequently occur during debugging. Here are the various forms
8579 @item @var{expression}
8580 Any expression valid in the current working language.
8582 @item @var{funcaddr}
8583 An address of a function or procedure derived from its name. In C,
8584 C@t{++}, Objective-C, Fortran, minimal, and assembly, this is
8585 simply the function's name @var{function} (and actually a special case
8586 of a valid expression). In Pascal and Modula-2, this is
8587 @code{&@var{function}}. In Ada, this is @code{@var{function}'Address}
8588 (although the Pascal form also works).
8590 This form specifies the address of the function's first instruction,
8591 before the stack frame and arguments have been set up.
8593 @item '@var{filename}':@var{funcaddr}
8594 Like @var{funcaddr} above, but also specifies the name of the source
8595 file explicitly. This is useful if the name of the function does not
8596 specify the function unambiguously, e.g., if there are several
8597 functions with identical names in different source files.
8601 @section Editing Source Files
8602 @cindex editing source files
8605 @kindex e @r{(@code{edit})}
8606 To edit the lines in a source file, use the @code{edit} command.
8607 The editing program of your choice
8608 is invoked with the current line set to
8609 the active line in the program.
8610 Alternatively, there are several ways to specify what part of the file you
8611 want to print if you want to see other parts of the program:
8614 @item edit @var{location}
8615 Edit the source file specified by @code{location}. Editing starts at
8616 that @var{location}, e.g., at the specified source line of the
8617 specified file. @xref{Specify Location}, for all the possible forms
8618 of the @var{location} argument; here are the forms of the @code{edit}
8619 command most commonly used:
8622 @item edit @var{number}
8623 Edit the current source file with @var{number} as the active line number.
8625 @item edit @var{function}
8626 Edit the file containing @var{function} at the beginning of its definition.
8631 @subsection Choosing your Editor
8632 You can customize @value{GDBN} to use any editor you want
8634 The only restriction is that your editor (say @code{ex}), recognizes the
8635 following command-line syntax:
8637 ex +@var{number} file
8639 The optional numeric value +@var{number} specifies the number of the line in
8640 the file where to start editing.}.
8641 By default, it is @file{@value{EDITOR}}, but you can change this
8642 by setting the environment variable @code{EDITOR} before using
8643 @value{GDBN}. For example, to configure @value{GDBN} to use the
8644 @code{vi} editor, you could use these commands with the @code{sh} shell:
8650 or in the @code{csh} shell,
8652 setenv EDITOR /usr/bin/vi
8657 @section Searching Source Files
8658 @cindex searching source files
8660 There are two commands for searching through the current source file for a
8665 @kindex forward-search
8666 @kindex fo @r{(@code{forward-search})}
8667 @item forward-search @var{regexp}
8668 @itemx search @var{regexp}
8669 The command @samp{forward-search @var{regexp}} checks each line,
8670 starting with the one following the last line listed, for a match for
8671 @var{regexp}. It lists the line that is found. You can use the
8672 synonym @samp{search @var{regexp}} or abbreviate the command name as
8675 @kindex reverse-search
8676 @item reverse-search @var{regexp}
8677 The command @samp{reverse-search @var{regexp}} checks each line, starting
8678 with the one before the last line listed and going backward, for a match
8679 for @var{regexp}. It lists the line that is found. You can abbreviate
8680 this command as @code{rev}.
8684 @section Specifying Source Directories
8687 @cindex directories for source files
8688 Executable programs sometimes do not record the directories of the source
8689 files from which they were compiled, just the names. Even when they do,
8690 the directories could be moved between the compilation and your debugging
8691 session. @value{GDBN} has a list of directories to search for source files;
8692 this is called the @dfn{source path}. Each time @value{GDBN} wants a source file,
8693 it tries all the directories in the list, in the order they are present
8694 in the list, until it finds a file with the desired name.
8696 For example, suppose an executable references the file
8697 @file{/usr/src/foo-1.0/lib/foo.c}, and our source path is
8698 @file{/mnt/cross}. The file is first looked up literally; if this
8699 fails, @file{/mnt/cross/usr/src/foo-1.0/lib/foo.c} is tried; if this
8700 fails, @file{/mnt/cross/foo.c} is opened; if this fails, an error
8701 message is printed. @value{GDBN} does not look up the parts of the
8702 source file name, such as @file{/mnt/cross/src/foo-1.0/lib/foo.c}.
8703 Likewise, the subdirectories of the source path are not searched: if
8704 the source path is @file{/mnt/cross}, and the binary refers to
8705 @file{foo.c}, @value{GDBN} would not find it under
8706 @file{/mnt/cross/usr/src/foo-1.0/lib}.
8708 Plain file names, relative file names with leading directories, file
8709 names containing dots, etc.@: are all treated as described above; for
8710 instance, if the source path is @file{/mnt/cross}, and the source file
8711 is recorded as @file{../lib/foo.c}, @value{GDBN} would first try
8712 @file{../lib/foo.c}, then @file{/mnt/cross/../lib/foo.c}, and after
8713 that---@file{/mnt/cross/foo.c}.
8715 Note that the executable search path is @emph{not} used to locate the
8718 Whenever you reset or rearrange the source path, @value{GDBN} clears out
8719 any information it has cached about where source files are found and where
8720 each line is in the file.
8724 When you start @value{GDBN}, its source path includes only @samp{cdir}
8725 and @samp{cwd}, in that order.
8726 To add other directories, use the @code{directory} command.
8728 The search path is used to find both program source files and @value{GDBN}
8729 script files (read using the @samp{-command} option and @samp{source} command).
8731 In addition to the source path, @value{GDBN} provides a set of commands
8732 that manage a list of source path substitution rules. A @dfn{substitution
8733 rule} specifies how to rewrite source directories stored in the program's
8734 debug information in case the sources were moved to a different
8735 directory between compilation and debugging. A rule is made of
8736 two strings, the first specifying what needs to be rewritten in
8737 the path, and the second specifying how it should be rewritten.
8738 In @ref{set substitute-path}, we name these two parts @var{from} and
8739 @var{to} respectively. @value{GDBN} does a simple string replacement
8740 of @var{from} with @var{to} at the start of the directory part of the
8741 source file name, and uses that result instead of the original file
8742 name to look up the sources.
8744 Using the previous example, suppose the @file{foo-1.0} tree has been
8745 moved from @file{/usr/src} to @file{/mnt/cross}, then you can tell
8746 @value{GDBN} to replace @file{/usr/src} in all source path names with
8747 @file{/mnt/cross}. The first lookup will then be
8748 @file{/mnt/cross/foo-1.0/lib/foo.c} in place of the original location
8749 of @file{/usr/src/foo-1.0/lib/foo.c}. To define a source path
8750 substitution rule, use the @code{set substitute-path} command
8751 (@pxref{set substitute-path}).
8753 To avoid unexpected substitution results, a rule is applied only if the
8754 @var{from} part of the directory name ends at a directory separator.
8755 For instance, a rule substituting @file{/usr/source} into
8756 @file{/mnt/cross} will be applied to @file{/usr/source/foo-1.0} but
8757 not to @file{/usr/sourceware/foo-2.0}. And because the substitution
8758 is applied only at the beginning of the directory name, this rule will
8759 not be applied to @file{/root/usr/source/baz.c} either.
8761 In many cases, you can achieve the same result using the @code{directory}
8762 command. However, @code{set substitute-path} can be more efficient in
8763 the case where the sources are organized in a complex tree with multiple
8764 subdirectories. With the @code{directory} command, you need to add each
8765 subdirectory of your project. If you moved the entire tree while
8766 preserving its internal organization, then @code{set substitute-path}
8767 allows you to direct the debugger to all the sources with one single
8770 @code{set substitute-path} is also more than just a shortcut command.
8771 The source path is only used if the file at the original location no
8772 longer exists. On the other hand, @code{set substitute-path} modifies
8773 the debugger behavior to look at the rewritten location instead. So, if
8774 for any reason a source file that is not relevant to your executable is
8775 located at the original location, a substitution rule is the only
8776 method available to point @value{GDBN} at the new location.
8778 @cindex @samp{--with-relocated-sources}
8779 @cindex default source path substitution
8780 You can configure a default source path substitution rule by
8781 configuring @value{GDBN} with the
8782 @samp{--with-relocated-sources=@var{dir}} option. The @var{dir}
8783 should be the name of a directory under @value{GDBN}'s configured
8784 prefix (set with @samp{--prefix} or @samp{--exec-prefix}), and
8785 directory names in debug information under @var{dir} will be adjusted
8786 automatically if the installed @value{GDBN} is moved to a new
8787 location. This is useful if @value{GDBN}, libraries or executables
8788 with debug information and corresponding source code are being moved
8792 @item directory @var{dirname} @dots{}
8793 @item dir @var{dirname} @dots{}
8794 Add directory @var{dirname} to the front of the source path. Several
8795 directory names may be given to this command, separated by @samp{:}
8796 (@samp{;} on MS-DOS and MS-Windows, where @samp{:} usually appears as
8797 part of absolute file names) or
8798 whitespace. You may specify a directory that is already in the source
8799 path; this moves it forward, so @value{GDBN} searches it sooner.
8803 @vindex $cdir@r{, convenience variable}
8804 @vindex $cwd@r{, convenience variable}
8805 @cindex compilation directory
8806 @cindex current directory
8807 @cindex working directory
8808 @cindex directory, current
8809 @cindex directory, compilation
8810 You can use the string @samp{$cdir} to refer to the compilation
8811 directory (if one is recorded), and @samp{$cwd} to refer to the current
8812 working directory. @samp{$cwd} is not the same as @samp{.}---the former
8813 tracks the current working directory as it changes during your @value{GDBN}
8814 session, while the latter is immediately expanded to the current
8815 directory at the time you add an entry to the source path.
8818 Reset the source path to its default value (@samp{$cdir:$cwd} on Unix systems). This requires confirmation.
8820 @c RET-repeat for @code{directory} is explicitly disabled, but since
8821 @c repeating it would be a no-op we do not say that. (thanks to RMS)
8823 @item set directories @var{path-list}
8824 @kindex set directories
8825 Set the source path to @var{path-list}.
8826 @samp{$cdir:$cwd} are added if missing.
8828 @item show directories
8829 @kindex show directories
8830 Print the source path: show which directories it contains.
8832 @anchor{set substitute-path}
8833 @item set substitute-path @var{from} @var{to}
8834 @kindex set substitute-path
8835 Define a source path substitution rule, and add it at the end of the
8836 current list of existing substitution rules. If a rule with the same
8837 @var{from} was already defined, then the old rule is also deleted.
8839 For example, if the file @file{/foo/bar/baz.c} was moved to
8840 @file{/mnt/cross/baz.c}, then the command
8843 (@value{GDBP}) set substitute-path /foo/bar /mnt/cross
8847 will tell @value{GDBN} to replace @samp{/foo/bar} with
8848 @samp{/mnt/cross}, which will allow @value{GDBN} to find the file
8849 @file{baz.c} even though it was moved.
8851 In the case when more than one substitution rule have been defined,
8852 the rules are evaluated one by one in the order where they have been
8853 defined. The first one matching, if any, is selected to perform
8856 For instance, if we had entered the following commands:
8859 (@value{GDBP}) set substitute-path /usr/src/include /mnt/include
8860 (@value{GDBP}) set substitute-path /usr/src /mnt/src
8864 @value{GDBN} would then rewrite @file{/usr/src/include/defs.h} into
8865 @file{/mnt/include/defs.h} by using the first rule. However, it would
8866 use the second rule to rewrite @file{/usr/src/lib/foo.c} into
8867 @file{/mnt/src/lib/foo.c}.
8870 @item unset substitute-path [path]
8871 @kindex unset substitute-path
8872 If a path is specified, search the current list of substitution rules
8873 for a rule that would rewrite that path. Delete that rule if found.
8874 A warning is emitted by the debugger if no rule could be found.
8876 If no path is specified, then all substitution rules are deleted.
8878 @item show substitute-path [path]
8879 @kindex show substitute-path
8880 If a path is specified, then print the source path substitution rule
8881 which would rewrite that path, if any.
8883 If no path is specified, then print all existing source path substitution
8888 If your source path is cluttered with directories that are no longer of
8889 interest, @value{GDBN} may sometimes cause confusion by finding the wrong
8890 versions of source. You can correct the situation as follows:
8894 Use @code{directory} with no argument to reset the source path to its default value.
8897 Use @code{directory} with suitable arguments to reinstall the
8898 directories you want in the source path. You can add all the
8899 directories in one command.
8903 @section Source and Machine Code
8904 @cindex source line and its code address
8906 You can use the command @code{info line} to map source lines to program
8907 addresses (and vice versa), and the command @code{disassemble} to display
8908 a range of addresses as machine instructions. You can use the command
8909 @code{set disassemble-next-line} to set whether to disassemble next
8910 source line when execution stops. When run under @sc{gnu} Emacs
8911 mode, the @code{info line} command causes the arrow to point to the
8912 line specified. Also, @code{info line} prints addresses in symbolic form as
8918 @itemx info line @var{location}
8919 Print the starting and ending addresses of the compiled code for
8920 source line @var{location}. You can specify source lines in any of
8921 the ways documented in @ref{Specify Location}. With no @var{location}
8922 information about the current source line is printed.
8925 For example, we can use @code{info line} to discover the location of
8926 the object code for the first line of function
8927 @code{m4_changequote}:
8930 (@value{GDBP}) info line m4_changequote
8931 Line 895 of "builtin.c" starts at pc 0x634c <m4_changequote> and \
8932 ends at 0x6350 <m4_changequote+4>.
8936 @cindex code address and its source line
8937 We can also inquire (using @code{*@var{addr}} as the form for
8938 @var{location}) what source line covers a particular address:
8940 (@value{GDBP}) info line *0x63ff
8941 Line 926 of "builtin.c" starts at pc 0x63e4 <m4_changequote+152> and \
8942 ends at 0x6404 <m4_changequote+184>.
8945 @cindex @code{$_} and @code{info line}
8946 @cindex @code{x} command, default address
8947 @kindex x@r{(examine), and} info line
8948 After @code{info line}, the default address for the @code{x} command
8949 is changed to the starting address of the line, so that @samp{x/i} is
8950 sufficient to begin examining the machine code (@pxref{Memory,
8951 ,Examining Memory}). Also, this address is saved as the value of the
8952 convenience variable @code{$_} (@pxref{Convenience Vars, ,Convenience
8955 @cindex info line, repeated calls
8956 After @code{info line}, using @code{info line} again without
8957 specifying a location will display information about the next source
8962 @cindex assembly instructions
8963 @cindex instructions, assembly
8964 @cindex machine instructions
8965 @cindex listing machine instructions
8967 @itemx disassemble /m
8968 @itemx disassemble /s
8969 @itemx disassemble /r
8970 This specialized command dumps a range of memory as machine
8971 instructions. It can also print mixed source+disassembly by specifying
8972 the @code{/m} or @code{/s} modifier and print the raw instructions in hex
8973 as well as in symbolic form by specifying the @code{/r} modifier.
8974 The default memory range is the function surrounding the
8975 program counter of the selected frame. A single argument to this
8976 command is a program counter value; @value{GDBN} dumps the function
8977 surrounding this value. When two arguments are given, they should
8978 be separated by a comma, possibly surrounded by whitespace. The
8979 arguments specify a range of addresses to dump, in one of two forms:
8982 @item @var{start},@var{end}
8983 the addresses from @var{start} (inclusive) to @var{end} (exclusive)
8984 @item @var{start},+@var{length}
8985 the addresses from @var{start} (inclusive) to
8986 @code{@var{start}+@var{length}} (exclusive).
8990 When 2 arguments are specified, the name of the function is also
8991 printed (since there could be several functions in the given range).
8993 The argument(s) can be any expression yielding a numeric value, such as
8994 @samp{0x32c4}, @samp{&main+10} or @samp{$pc - 8}.
8996 If the range of memory being disassembled contains current program counter,
8997 the instruction at that location is shown with a @code{=>} marker.
9000 The following example shows the disassembly of a range of addresses of
9001 HP PA-RISC 2.0 code:
9004 (@value{GDBP}) disas 0x32c4, 0x32e4
9005 Dump of assembler code from 0x32c4 to 0x32e4:
9006 0x32c4 <main+204>: addil 0,dp
9007 0x32c8 <main+208>: ldw 0x22c(sr0,r1),r26
9008 0x32cc <main+212>: ldil 0x3000,r31
9009 0x32d0 <main+216>: ble 0x3f8(sr4,r31)
9010 0x32d4 <main+220>: ldo 0(r31),rp
9011 0x32d8 <main+224>: addil -0x800,dp
9012 0x32dc <main+228>: ldo 0x588(r1),r26
9013 0x32e0 <main+232>: ldil 0x3000,r31
9014 End of assembler dump.
9017 Here is an example showing mixed source+assembly for Intel x86
9018 with @code{/m} or @code{/s}, when the program is stopped just after
9019 function prologue in a non-optimized function with no inline code.
9022 (@value{GDBP}) disas /m main
9023 Dump of assembler code for function main:
9025 0x08048330 <+0>: push %ebp
9026 0x08048331 <+1>: mov %esp,%ebp
9027 0x08048333 <+3>: sub $0x8,%esp
9028 0x08048336 <+6>: and $0xfffffff0,%esp
9029 0x08048339 <+9>: sub $0x10,%esp
9031 6 printf ("Hello.\n");
9032 => 0x0804833c <+12>: movl $0x8048440,(%esp)
9033 0x08048343 <+19>: call 0x8048284 <puts@@plt>
9037 0x08048348 <+24>: mov $0x0,%eax
9038 0x0804834d <+29>: leave
9039 0x0804834e <+30>: ret
9041 End of assembler dump.
9044 The @code{/m} option is deprecated as its output is not useful when
9045 there is either inlined code or re-ordered code.
9046 The @code{/s} option is the preferred choice.
9047 Here is an example for AMD x86-64 showing the difference between
9048 @code{/m} output and @code{/s} output.
9049 This example has one inline function defined in a header file,
9050 and the code is compiled with @samp{-O2} optimization.
9051 Note how the @code{/m} output is missing the disassembly of
9052 several instructions that are present in the @code{/s} output.
9082 (@value{GDBP}) disas /m main
9083 Dump of assembler code for function main:
9087 0x0000000000400400 <+0>: mov 0x200c2e(%rip),%eax # 0x601034 <y>
9088 0x0000000000400417 <+23>: mov %eax,0x200c13(%rip) # 0x601030 <x>
9092 0x000000000040041d <+29>: xor %eax,%eax
9093 0x000000000040041f <+31>: retq
9094 0x0000000000400420 <+32>: add %eax,%eax
9095 0x0000000000400422 <+34>: jmp 0x400417 <main+23>
9097 End of assembler dump.
9098 (@value{GDBP}) disas /s main
9099 Dump of assembler code for function main:
9103 0x0000000000400400 <+0>: mov 0x200c2e(%rip),%eax # 0x601034 <y>
9107 0x0000000000400406 <+6>: test %eax,%eax
9108 0x0000000000400408 <+8>: js 0x400420 <main+32>
9113 0x000000000040040a <+10>: lea 0xa(%rax),%edx
9114 0x000000000040040d <+13>: test %eax,%eax
9115 0x000000000040040f <+15>: mov $0x1,%eax
9116 0x0000000000400414 <+20>: cmovne %edx,%eax
9120 0x0000000000400417 <+23>: mov %eax,0x200c13(%rip) # 0x601030 <x>
9124 0x000000000040041d <+29>: xor %eax,%eax
9125 0x000000000040041f <+31>: retq
9129 0x0000000000400420 <+32>: add %eax,%eax
9130 0x0000000000400422 <+34>: jmp 0x400417 <main+23>
9131 End of assembler dump.
9134 Here is another example showing raw instructions in hex for AMD x86-64,
9137 (gdb) disas /r 0x400281,+10
9138 Dump of assembler code from 0x400281 to 0x40028b:
9139 0x0000000000400281: 38 36 cmp %dh,(%rsi)
9140 0x0000000000400283: 2d 36 34 2e 73 sub $0x732e3436,%eax
9141 0x0000000000400288: 6f outsl %ds:(%rsi),(%dx)
9142 0x0000000000400289: 2e 32 00 xor %cs:(%rax),%al
9143 End of assembler dump.
9146 Addresses cannot be specified as a location (@pxref{Specify Location}).
9147 So, for example, if you want to disassemble function @code{bar}
9148 in file @file{foo.c}, you must type @samp{disassemble 'foo.c'::bar}
9149 and not @samp{disassemble foo.c:bar}.
9151 Some architectures have more than one commonly-used set of instruction
9152 mnemonics or other syntax.
9154 For programs that were dynamically linked and use shared libraries,
9155 instructions that call functions or branch to locations in the shared
9156 libraries might show a seemingly bogus location---it's actually a
9157 location of the relocation table. On some architectures, @value{GDBN}
9158 might be able to resolve these to actual function names.
9161 @kindex set disassembler-options
9162 @cindex disassembler options
9163 @item set disassembler-options @var{option1}[,@var{option2}@dots{}]
9164 This command controls the passing of target specific information to
9165 the disassembler. For a list of valid options, please refer to the
9166 @code{-M}/@code{--disassembler-options} section of the @samp{objdump}
9167 manual and/or the output of @kbd{objdump --help}
9168 (@pxref{objdump,,objdump,binutils,The GNU Binary Utilities}).
9169 The default value is the empty string.
9171 If it is necessary to specify more than one disassembler option, then
9172 multiple options can be placed together into a comma separated list.
9173 Currently this command is only supported on targets ARM, MIPS, PowerPC
9176 @kindex show disassembler-options
9177 @item show disassembler-options
9178 Show the current setting of the disassembler options.
9182 @kindex set disassembly-flavor
9183 @cindex Intel disassembly flavor
9184 @cindex AT&T disassembly flavor
9185 @item set disassembly-flavor @var{instruction-set}
9186 Select the instruction set to use when disassembling the
9187 program via the @code{disassemble} or @code{x/i} commands.
9189 Currently this command is only defined for the Intel x86 family. You
9190 can set @var{instruction-set} to either @code{intel} or @code{att}.
9191 The default is @code{att}, the AT&T flavor used by default by Unix
9192 assemblers for x86-based targets.
9194 @kindex show disassembly-flavor
9195 @item show disassembly-flavor
9196 Show the current setting of the disassembly flavor.
9200 @kindex set disassemble-next-line
9201 @kindex show disassemble-next-line
9202 @item set disassemble-next-line
9203 @itemx show disassemble-next-line
9204 Control whether or not @value{GDBN} will disassemble the next source
9205 line or instruction when execution stops. If ON, @value{GDBN} will
9206 display disassembly of the next source line when execution of the
9207 program being debugged stops. This is @emph{in addition} to
9208 displaying the source line itself, which @value{GDBN} always does if
9209 possible. If the next source line cannot be displayed for some reason
9210 (e.g., if @value{GDBN} cannot find the source file, or there's no line
9211 info in the debug info), @value{GDBN} will display disassembly of the
9212 next @emph{instruction} instead of showing the next source line. If
9213 AUTO, @value{GDBN} will display disassembly of next instruction only
9214 if the source line cannot be displayed. This setting causes
9215 @value{GDBN} to display some feedback when you step through a function
9216 with no line info or whose source file is unavailable. The default is
9217 OFF, which means never display the disassembly of the next line or
9223 @chapter Examining Data
9225 @cindex printing data
9226 @cindex examining data
9229 The usual way to examine data in your program is with the @code{print}
9230 command (abbreviated @code{p}), or its synonym @code{inspect}. It
9231 evaluates and prints the value of an expression of the language your
9232 program is written in (@pxref{Languages, ,Using @value{GDBN} with
9233 Different Languages}). It may also print the expression using a
9234 Python-based pretty-printer (@pxref{Pretty Printing}).
9237 @item print @var{expr}
9238 @itemx print /@var{f} @var{expr}
9239 @var{expr} is an expression (in the source language). By default the
9240 value of @var{expr} is printed in a format appropriate to its data type;
9241 you can choose a different format by specifying @samp{/@var{f}}, where
9242 @var{f} is a letter specifying the format; see @ref{Output Formats,,Output
9246 @itemx print /@var{f}
9247 @cindex reprint the last value
9248 If you omit @var{expr}, @value{GDBN} displays the last value again (from the
9249 @dfn{value history}; @pxref{Value History, ,Value History}). This allows you to
9250 conveniently inspect the same value in an alternative format.
9253 A more low-level way of examining data is with the @code{x} command.
9254 It examines data in memory at a specified address and prints it in a
9255 specified format. @xref{Memory, ,Examining Memory}.
9257 If you are interested in information about types, or about how the
9258 fields of a struct or a class are declared, use the @code{ptype @var{exp}}
9259 command rather than @code{print}. @xref{Symbols, ,Examining the Symbol
9262 @cindex exploring hierarchical data structures
9264 Another way of examining values of expressions and type information is
9265 through the Python extension command @code{explore} (available only if
9266 the @value{GDBN} build is configured with @code{--with-python}). It
9267 offers an interactive way to start at the highest level (or, the most
9268 abstract level) of the data type of an expression (or, the data type
9269 itself) and explore all the way down to leaf scalar values/fields
9270 embedded in the higher level data types.
9273 @item explore @var{arg}
9274 @var{arg} is either an expression (in the source language), or a type
9275 visible in the current context of the program being debugged.
9278 The working of the @code{explore} command can be illustrated with an
9279 example. If a data type @code{struct ComplexStruct} is defined in your
9289 struct ComplexStruct
9291 struct SimpleStruct *ss_p;
9297 followed by variable declarations as
9300 struct SimpleStruct ss = @{ 10, 1.11 @};
9301 struct ComplexStruct cs = @{ &ss, @{ 0, 1, 2, 3, 4, 5, 6, 7, 8, 9 @} @};
9305 then, the value of the variable @code{cs} can be explored using the
9306 @code{explore} command as follows.
9310 The value of `cs' is a struct/class of type `struct ComplexStruct' with
9311 the following fields:
9313 ss_p = <Enter 0 to explore this field of type `struct SimpleStruct *'>
9314 arr = <Enter 1 to explore this field of type `int [10]'>
9316 Enter the field number of choice:
9320 Since the fields of @code{cs} are not scalar values, you are being
9321 prompted to chose the field you want to explore. Let's say you choose
9322 the field @code{ss_p} by entering @code{0}. Then, since this field is a
9323 pointer, you will be asked if it is pointing to a single value. From
9324 the declaration of @code{cs} above, it is indeed pointing to a single
9325 value, hence you enter @code{y}. If you enter @code{n}, then you will
9326 be asked if it were pointing to an array of values, in which case this
9327 field will be explored as if it were an array.
9330 `cs.ss_p' is a pointer to a value of type `struct SimpleStruct'
9331 Continue exploring it as a pointer to a single value [y/n]: y
9332 The value of `*(cs.ss_p)' is a struct/class of type `struct
9333 SimpleStruct' with the following fields:
9335 i = 10 .. (Value of type `int')
9336 d = 1.1100000000000001 .. (Value of type `double')
9338 Press enter to return to parent value:
9342 If the field @code{arr} of @code{cs} was chosen for exploration by
9343 entering @code{1} earlier, then since it is as array, you will be
9344 prompted to enter the index of the element in the array that you want
9348 `cs.arr' is an array of `int'.
9349 Enter the index of the element you want to explore in `cs.arr': 5
9351 `(cs.arr)[5]' is a scalar value of type `int'.
9355 Press enter to return to parent value:
9358 In general, at any stage of exploration, you can go deeper towards the
9359 leaf values by responding to the prompts appropriately, or hit the
9360 return key to return to the enclosing data structure (the @i{higher}
9361 level data structure).
9363 Similar to exploring values, you can use the @code{explore} command to
9364 explore types. Instead of specifying a value (which is typically a
9365 variable name or an expression valid in the current context of the
9366 program being debugged), you specify a type name. If you consider the
9367 same example as above, your can explore the type
9368 @code{struct ComplexStruct} by passing the argument
9369 @code{struct ComplexStruct} to the @code{explore} command.
9372 (gdb) explore struct ComplexStruct
9376 By responding to the prompts appropriately in the subsequent interactive
9377 session, you can explore the type @code{struct ComplexStruct} in a
9378 manner similar to how the value @code{cs} was explored in the above
9381 The @code{explore} command also has two sub-commands,
9382 @code{explore value} and @code{explore type}. The former sub-command is
9383 a way to explicitly specify that value exploration of the argument is
9384 being invoked, while the latter is a way to explicitly specify that type
9385 exploration of the argument is being invoked.
9388 @item explore value @var{expr}
9389 @cindex explore value
9390 This sub-command of @code{explore} explores the value of the
9391 expression @var{expr} (if @var{expr} is an expression valid in the
9392 current context of the program being debugged). The behavior of this
9393 command is identical to that of the behavior of the @code{explore}
9394 command being passed the argument @var{expr}.
9396 @item explore type @var{arg}
9397 @cindex explore type
9398 This sub-command of @code{explore} explores the type of @var{arg} (if
9399 @var{arg} is a type visible in the current context of program being
9400 debugged), or the type of the value/expression @var{arg} (if @var{arg}
9401 is an expression valid in the current context of the program being
9402 debugged). If @var{arg} is a type, then the behavior of this command is
9403 identical to that of the @code{explore} command being passed the
9404 argument @var{arg}. If @var{arg} is an expression, then the behavior of
9405 this command will be identical to that of the @code{explore} command
9406 being passed the type of @var{arg} as the argument.
9410 * Expressions:: Expressions
9411 * Ambiguous Expressions:: Ambiguous Expressions
9412 * Variables:: Program variables
9413 * Arrays:: Artificial arrays
9414 * Output Formats:: Output formats
9415 * Memory:: Examining memory
9416 * Auto Display:: Automatic display
9417 * Print Settings:: Print settings
9418 * Pretty Printing:: Python pretty printing
9419 * Value History:: Value history
9420 * Convenience Vars:: Convenience variables
9421 * Convenience Funs:: Convenience functions
9422 * Registers:: Registers
9423 * Floating Point Hardware:: Floating point hardware
9424 * Vector Unit:: Vector Unit
9425 * OS Information:: Auxiliary data provided by operating system
9426 * Memory Region Attributes:: Memory region attributes
9427 * Dump/Restore Files:: Copy between memory and a file
9428 * Core File Generation:: Cause a program dump its core
9429 * Character Sets:: Debugging programs that use a different
9430 character set than GDB does
9431 * Caching Target Data:: Data caching for targets
9432 * Searching Memory:: Searching memory for a sequence of bytes
9433 * Value Sizes:: Managing memory allocated for values
9437 @section Expressions
9440 @code{print} and many other @value{GDBN} commands accept an expression and
9441 compute its value. Any kind of constant, variable or operator defined
9442 by the programming language you are using is valid in an expression in
9443 @value{GDBN}. This includes conditional expressions, function calls,
9444 casts, and string constants. It also includes preprocessor macros, if
9445 you compiled your program to include this information; see
9448 @cindex arrays in expressions
9449 @value{GDBN} supports array constants in expressions input by
9450 the user. The syntax is @{@var{element}, @var{element}@dots{}@}. For example,
9451 you can use the command @code{print @{1, 2, 3@}} to create an array
9452 of three integers. If you pass an array to a function or assign it
9453 to a program variable, @value{GDBN} copies the array to memory that
9454 is @code{malloc}ed in the target program.
9456 Because C is so widespread, most of the expressions shown in examples in
9457 this manual are in C. @xref{Languages, , Using @value{GDBN} with Different
9458 Languages}, for information on how to use expressions in other
9461 In this section, we discuss operators that you can use in @value{GDBN}
9462 expressions regardless of your programming language.
9464 @cindex casts, in expressions
9465 Casts are supported in all languages, not just in C, because it is so
9466 useful to cast a number into a pointer in order to examine a structure
9467 at that address in memory.
9468 @c FIXME: casts supported---Mod2 true?
9470 @value{GDBN} supports these operators, in addition to those common
9471 to programming languages:
9475 @samp{@@} is a binary operator for treating parts of memory as arrays.
9476 @xref{Arrays, ,Artificial Arrays}, for more information.
9479 @samp{::} allows you to specify a variable in terms of the file or
9480 function where it is defined. @xref{Variables, ,Program Variables}.
9482 @cindex @{@var{type}@}
9483 @cindex type casting memory
9484 @cindex memory, viewing as typed object
9485 @cindex casts, to view memory
9486 @item @{@var{type}@} @var{addr}
9487 Refers to an object of type @var{type} stored at address @var{addr} in
9488 memory. The address @var{addr} may be any expression whose value is
9489 an integer or pointer (but parentheses are required around binary
9490 operators, just as in a cast). This construct is allowed regardless
9491 of what kind of data is normally supposed to reside at @var{addr}.
9494 @node Ambiguous Expressions
9495 @section Ambiguous Expressions
9496 @cindex ambiguous expressions
9498 Expressions can sometimes contain some ambiguous elements. For instance,
9499 some programming languages (notably Ada, C@t{++} and Objective-C) permit
9500 a single function name to be defined several times, for application in
9501 different contexts. This is called @dfn{overloading}. Another example
9502 involving Ada is generics. A @dfn{generic package} is similar to C@t{++}
9503 templates and is typically instantiated several times, resulting in
9504 the same function name being defined in different contexts.
9506 In some cases and depending on the language, it is possible to adjust
9507 the expression to remove the ambiguity. For instance in C@t{++}, you
9508 can specify the signature of the function you want to break on, as in
9509 @kbd{break @var{function}(@var{types})}. In Ada, using the fully
9510 qualified name of your function often makes the expression unambiguous
9513 When an ambiguity that needs to be resolved is detected, the debugger
9514 has the capability to display a menu of numbered choices for each
9515 possibility, and then waits for the selection with the prompt @samp{>}.
9516 The first option is always @samp{[0] cancel}, and typing @kbd{0 @key{RET}}
9517 aborts the current command. If the command in which the expression was
9518 used allows more than one choice to be selected, the next option in the
9519 menu is @samp{[1] all}, and typing @kbd{1 @key{RET}} selects all possible
9522 For example, the following session excerpt shows an attempt to set a
9523 breakpoint at the overloaded symbol @code{String::after}.
9524 We choose three particular definitions of that function name:
9526 @c FIXME! This is likely to change to show arg type lists, at least
9529 (@value{GDBP}) b String::after
9532 [2] file:String.cc; line number:867
9533 [3] file:String.cc; line number:860
9534 [4] file:String.cc; line number:875
9535 [5] file:String.cc; line number:853
9536 [6] file:String.cc; line number:846
9537 [7] file:String.cc; line number:735
9539 Breakpoint 1 at 0xb26c: file String.cc, line 867.
9540 Breakpoint 2 at 0xb344: file String.cc, line 875.
9541 Breakpoint 3 at 0xafcc: file String.cc, line 846.
9542 Multiple breakpoints were set.
9543 Use the "delete" command to delete unwanted
9550 @kindex set multiple-symbols
9551 @item set multiple-symbols @var{mode}
9552 @cindex multiple-symbols menu
9554 This option allows you to adjust the debugger behavior when an expression
9557 By default, @var{mode} is set to @code{all}. If the command with which
9558 the expression is used allows more than one choice, then @value{GDBN}
9559 automatically selects all possible choices. For instance, inserting
9560 a breakpoint on a function using an ambiguous name results in a breakpoint
9561 inserted on each possible match. However, if a unique choice must be made,
9562 then @value{GDBN} uses the menu to help you disambiguate the expression.
9563 For instance, printing the address of an overloaded function will result
9564 in the use of the menu.
9566 When @var{mode} is set to @code{ask}, the debugger always uses the menu
9567 when an ambiguity is detected.
9569 Finally, when @var{mode} is set to @code{cancel}, the debugger reports
9570 an error due to the ambiguity and the command is aborted.
9572 @kindex show multiple-symbols
9573 @item show multiple-symbols
9574 Show the current value of the @code{multiple-symbols} setting.
9578 @section Program Variables
9580 The most common kind of expression to use is the name of a variable
9583 Variables in expressions are understood in the selected stack frame
9584 (@pxref{Selection, ,Selecting a Frame}); they must be either:
9588 global (or file-static)
9595 visible according to the scope rules of the
9596 programming language from the point of execution in that frame
9599 @noindent This means that in the function
9614 you can examine and use the variable @code{a} whenever your program is
9615 executing within the function @code{foo}, but you can only use or
9616 examine the variable @code{b} while your program is executing inside
9617 the block where @code{b} is declared.
9619 @cindex variable name conflict
9620 There is an exception: you can refer to a variable or function whose
9621 scope is a single source file even if the current execution point is not
9622 in this file. But it is possible to have more than one such variable or
9623 function with the same name (in different source files). If that
9624 happens, referring to that name has unpredictable effects. If you wish,
9625 you can specify a static variable in a particular function or file by
9626 using the colon-colon (@code{::}) notation:
9628 @cindex colon-colon, context for variables/functions
9630 @c info cannot cope with a :: index entry, but why deprive hard copy readers?
9631 @cindex @code{::}, context for variables/functions
9634 @var{file}::@var{variable}
9635 @var{function}::@var{variable}
9639 Here @var{file} or @var{function} is the name of the context for the
9640 static @var{variable}. In the case of file names, you can use quotes to
9641 make sure @value{GDBN} parses the file name as a single word---for example,
9642 to print a global value of @code{x} defined in @file{f2.c}:
9645 (@value{GDBP}) p 'f2.c'::x
9648 The @code{::} notation is normally used for referring to
9649 static variables, since you typically disambiguate uses of local variables
9650 in functions by selecting the appropriate frame and using the
9651 simple name of the variable. However, you may also use this notation
9652 to refer to local variables in frames enclosing the selected frame:
9661 process (a); /* Stop here */
9672 For example, if there is a breakpoint at the commented line,
9673 here is what you might see
9674 when the program stops after executing the call @code{bar(0)}:
9679 (@value{GDBP}) p bar::a
9682 #2 0x080483d0 in foo (a=5) at foobar.c:12
9685 (@value{GDBP}) p bar::a
9689 @cindex C@t{++} scope resolution
9690 These uses of @samp{::} are very rarely in conflict with the very
9691 similar use of the same notation in C@t{++}. When they are in
9692 conflict, the C@t{++} meaning takes precedence; however, this can be
9693 overridden by quoting the file or function name with single quotes.
9695 For example, suppose the program is stopped in a method of a class
9696 that has a field named @code{includefile}, and there is also an
9697 include file named @file{includefile} that defines a variable,
9701 (@value{GDBP}) p includefile
9703 (@value{GDBP}) p includefile::some_global
9704 A syntax error in expression, near `'.
9705 (@value{GDBP}) p 'includefile'::some_global
9709 @cindex wrong values
9710 @cindex variable values, wrong
9711 @cindex function entry/exit, wrong values of variables
9712 @cindex optimized code, wrong values of variables
9714 @emph{Warning:} Occasionally, a local variable may appear to have the
9715 wrong value at certain points in a function---just after entry to a new
9716 scope, and just before exit.
9718 You may see this problem when you are stepping by machine instructions.
9719 This is because, on most machines, it takes more than one instruction to
9720 set up a stack frame (including local variable definitions); if you are
9721 stepping by machine instructions, variables may appear to have the wrong
9722 values until the stack frame is completely built. On exit, it usually
9723 also takes more than one machine instruction to destroy a stack frame;
9724 after you begin stepping through that group of instructions, local
9725 variable definitions may be gone.
9727 This may also happen when the compiler does significant optimizations.
9728 To be sure of always seeing accurate values, turn off all optimization
9731 @cindex ``No symbol "foo" in current context''
9732 Another possible effect of compiler optimizations is to optimize
9733 unused variables out of existence, or assign variables to registers (as
9734 opposed to memory addresses). Depending on the support for such cases
9735 offered by the debug info format used by the compiler, @value{GDBN}
9736 might not be able to display values for such local variables. If that
9737 happens, @value{GDBN} will print a message like this:
9740 No symbol "foo" in current context.
9743 To solve such problems, either recompile without optimizations, or use a
9744 different debug info format, if the compiler supports several such
9745 formats. @xref{Compilation}, for more information on choosing compiler
9746 options. @xref{C, ,C and C@t{++}}, for more information about debug
9747 info formats that are best suited to C@t{++} programs.
9749 If you ask to print an object whose contents are unknown to
9750 @value{GDBN}, e.g., because its data type is not completely specified
9751 by the debug information, @value{GDBN} will say @samp{<incomplete
9752 type>}. @xref{Symbols, incomplete type}, for more about this.
9754 @cindex no debug info variables
9755 If you try to examine or use the value of a (global) variable for
9756 which @value{GDBN} has no type information, e.g., because the program
9757 includes no debug information, @value{GDBN} displays an error message.
9758 @xref{Symbols, unknown type}, for more about unknown types. If you
9759 cast the variable to its declared type, @value{GDBN} gets the
9760 variable's value using the cast-to type as the variable's type. For
9761 example, in a C program:
9764 (@value{GDBP}) p var
9765 'var' has unknown type; cast it to its declared type
9766 (@value{GDBP}) p (float) var
9770 If you append @kbd{@@entry} string to a function parameter name you get its
9771 value at the time the function got called. If the value is not available an
9772 error message is printed. Entry values are available only with some compilers.
9773 Entry values are normally also printed at the function parameter list according
9774 to @ref{set print entry-values}.
9777 Breakpoint 1, d (i=30) at gdb.base/entry-value.c:29
9783 (gdb) print i@@entry
9787 Strings are identified as arrays of @code{char} values without specified
9788 signedness. Arrays of either @code{signed char} or @code{unsigned char} get
9789 printed as arrays of 1 byte sized integers. @code{-fsigned-char} or
9790 @code{-funsigned-char} @value{NGCC} options have no effect as @value{GDBN}
9791 defines literal string type @code{"char"} as @code{char} without a sign.
9796 signed char var1[] = "A";
9799 You get during debugging
9804 $2 = @{65 'A', 0 '\0'@}
9808 @section Artificial Arrays
9810 @cindex artificial array
9812 @kindex @@@r{, referencing memory as an array}
9813 It is often useful to print out several successive objects of the
9814 same type in memory; a section of an array, or an array of
9815 dynamically determined size for which only a pointer exists in the
9818 You can do this by referring to a contiguous span of memory as an
9819 @dfn{artificial array}, using the binary operator @samp{@@}. The left
9820 operand of @samp{@@} should be the first element of the desired array
9821 and be an individual object. The right operand should be the desired length
9822 of the array. The result is an array value whose elements are all of
9823 the type of the left argument. The first element is actually the left
9824 argument; the second element comes from bytes of memory immediately
9825 following those that hold the first element, and so on. Here is an
9826 example. If a program says
9829 int *array = (int *) malloc (len * sizeof (int));
9833 you can print the contents of @code{array} with
9839 The left operand of @samp{@@} must reside in memory. Array values made
9840 with @samp{@@} in this way behave just like other arrays in terms of
9841 subscripting, and are coerced to pointers when used in expressions.
9842 Artificial arrays most often appear in expressions via the value history
9843 (@pxref{Value History, ,Value History}), after printing one out.
9845 Another way to create an artificial array is to use a cast.
9846 This re-interprets a value as if it were an array.
9847 The value need not be in memory:
9849 (@value{GDBP}) p/x (short[2])0x12345678
9850 $1 = @{0x1234, 0x5678@}
9853 As a convenience, if you leave the array length out (as in
9854 @samp{(@var{type}[])@var{value}}) @value{GDBN} calculates the size to fill
9855 the value (as @samp{sizeof(@var{value})/sizeof(@var{type})}:
9857 (@value{GDBP}) p/x (short[])0x12345678
9858 $2 = @{0x1234, 0x5678@}
9861 Sometimes the artificial array mechanism is not quite enough; in
9862 moderately complex data structures, the elements of interest may not
9863 actually be adjacent---for example, if you are interested in the values
9864 of pointers in an array. One useful work-around in this situation is
9865 to use a convenience variable (@pxref{Convenience Vars, ,Convenience
9866 Variables}) as a counter in an expression that prints the first
9867 interesting value, and then repeat that expression via @key{RET}. For
9868 instance, suppose you have an array @code{dtab} of pointers to
9869 structures, and you are interested in the values of a field @code{fv}
9870 in each structure. Here is an example of what you might type:
9880 @node Output Formats
9881 @section Output Formats
9883 @cindex formatted output
9884 @cindex output formats
9885 By default, @value{GDBN} prints a value according to its data type. Sometimes
9886 this is not what you want. For example, you might want to print a number
9887 in hex, or a pointer in decimal. Or you might want to view data in memory
9888 at a certain address as a character string or as an instruction. To do
9889 these things, specify an @dfn{output format} when you print a value.
9891 The simplest use of output formats is to say how to print a value
9892 already computed. This is done by starting the arguments of the
9893 @code{print} command with a slash and a format letter. The format
9894 letters supported are:
9898 Regard the bits of the value as an integer, and print the integer in
9902 Print as integer in signed decimal.
9905 Print as integer in unsigned decimal.
9908 Print as integer in octal.
9911 Print as integer in binary. The letter @samp{t} stands for ``two''.
9912 @footnote{@samp{b} cannot be used because these format letters are also
9913 used with the @code{x} command, where @samp{b} stands for ``byte'';
9914 see @ref{Memory,,Examining Memory}.}
9917 @cindex unknown address, locating
9918 @cindex locate address
9919 Print as an address, both absolute in hexadecimal and as an offset from
9920 the nearest preceding symbol. You can use this format used to discover
9921 where (in what function) an unknown address is located:
9924 (@value{GDBP}) p/a 0x54320
9925 $3 = 0x54320 <_initialize_vx+396>
9929 The command @code{info symbol 0x54320} yields similar results.
9930 @xref{Symbols, info symbol}.
9933 Regard as an integer and print it as a character constant. This
9934 prints both the numerical value and its character representation. The
9935 character representation is replaced with the octal escape @samp{\nnn}
9936 for characters outside the 7-bit @sc{ascii} range.
9938 Without this format, @value{GDBN} displays @code{char},
9939 @w{@code{unsigned char}}, and @w{@code{signed char}} data as character
9940 constants. Single-byte members of vectors are displayed as integer
9944 Regard the bits of the value as a floating point number and print
9945 using typical floating point syntax.
9948 @cindex printing strings
9949 @cindex printing byte arrays
9950 Regard as a string, if possible. With this format, pointers to single-byte
9951 data are displayed as null-terminated strings and arrays of single-byte data
9952 are displayed as fixed-length strings. Other values are displayed in their
9955 Without this format, @value{GDBN} displays pointers to and arrays of
9956 @code{char}, @w{@code{unsigned char}}, and @w{@code{signed char}} as
9957 strings. Single-byte members of a vector are displayed as an integer
9961 Like @samp{x} formatting, the value is treated as an integer and
9962 printed as hexadecimal, but leading zeros are printed to pad the value
9963 to the size of the integer type.
9966 @cindex raw printing
9967 Print using the @samp{raw} formatting. By default, @value{GDBN} will
9968 use a Python-based pretty-printer, if one is available (@pxref{Pretty
9969 Printing}). This typically results in a higher-level display of the
9970 value's contents. The @samp{r} format bypasses any Python
9971 pretty-printer which might exist.
9974 For example, to print the program counter in hex (@pxref{Registers}), type
9981 Note that no space is required before the slash; this is because command
9982 names in @value{GDBN} cannot contain a slash.
9984 To reprint the last value in the value history with a different format,
9985 you can use the @code{print} command with just a format and no
9986 expression. For example, @samp{p/x} reprints the last value in hex.
9989 @section Examining Memory
9991 You can use the command @code{x} (for ``examine'') to examine memory in
9992 any of several formats, independently of your program's data types.
9994 @cindex examining memory
9996 @kindex x @r{(examine memory)}
9997 @item x/@var{nfu} @var{addr}
10000 Use the @code{x} command to examine memory.
10003 @var{n}, @var{f}, and @var{u} are all optional parameters that specify how
10004 much memory to display and how to format it; @var{addr} is an
10005 expression giving the address where you want to start displaying memory.
10006 If you use defaults for @var{nfu}, you need not type the slash @samp{/}.
10007 Several commands set convenient defaults for @var{addr}.
10010 @item @var{n}, the repeat count
10011 The repeat count is a decimal integer; the default is 1. It specifies
10012 how much memory (counting by units @var{u}) to display. If a negative
10013 number is specified, memory is examined backward from @var{addr}.
10014 @c This really is **decimal**; unaffected by 'set radix' as of GDB
10017 @item @var{f}, the display format
10018 The display format is one of the formats used by @code{print}
10019 (@samp{x}, @samp{d}, @samp{u}, @samp{o}, @samp{t}, @samp{a}, @samp{c},
10020 @samp{f}, @samp{s}), and in addition @samp{i} (for machine instructions).
10021 The default is @samp{x} (hexadecimal) initially. The default changes
10022 each time you use either @code{x} or @code{print}.
10024 @item @var{u}, the unit size
10025 The unit size is any of
10031 Halfwords (two bytes).
10033 Words (four bytes). This is the initial default.
10035 Giant words (eight bytes).
10038 Each time you specify a unit size with @code{x}, that size becomes the
10039 default unit the next time you use @code{x}. For the @samp{i} format,
10040 the unit size is ignored and is normally not written. For the @samp{s} format,
10041 the unit size defaults to @samp{b}, unless it is explicitly given.
10042 Use @kbd{x /hs} to display 16-bit char strings and @kbd{x /ws} to display
10043 32-bit strings. The next use of @kbd{x /s} will again display 8-bit strings.
10044 Note that the results depend on the programming language of the
10045 current compilation unit. If the language is C, the @samp{s}
10046 modifier will use the UTF-16 encoding while @samp{w} will use
10047 UTF-32. The encoding is set by the programming language and cannot
10050 @item @var{addr}, starting display address
10051 @var{addr} is the address where you want @value{GDBN} to begin displaying
10052 memory. The expression need not have a pointer value (though it may);
10053 it is always interpreted as an integer address of a byte of memory.
10054 @xref{Expressions, ,Expressions}, for more information on expressions. The default for
10055 @var{addr} is usually just after the last address examined---but several
10056 other commands also set the default address: @code{info breakpoints} (to
10057 the address of the last breakpoint listed), @code{info line} (to the
10058 starting address of a line), and @code{print} (if you use it to display
10059 a value from memory).
10062 For example, @samp{x/3uh 0x54320} is a request to display three halfwords
10063 (@code{h}) of memory, formatted as unsigned decimal integers (@samp{u}),
10064 starting at address @code{0x54320}. @samp{x/4xw $sp} prints the four
10065 words (@samp{w}) of memory above the stack pointer (here, @samp{$sp};
10066 @pxref{Registers, ,Registers}) in hexadecimal (@samp{x}).
10068 You can also specify a negative repeat count to examine memory backward
10069 from the given address. For example, @samp{x/-3uh 0x54320} prints three
10070 halfwords (@code{h}) at @code{0x54314}, @code{0x54328}, and @code{0x5431c}.
10072 Since the letters indicating unit sizes are all distinct from the
10073 letters specifying output formats, you do not have to remember whether
10074 unit size or format comes first; either order works. The output
10075 specifications @samp{4xw} and @samp{4wx} mean exactly the same thing.
10076 (However, the count @var{n} must come first; @samp{wx4} does not work.)
10078 Even though the unit size @var{u} is ignored for the formats @samp{s}
10079 and @samp{i}, you might still want to use a count @var{n}; for example,
10080 @samp{3i} specifies that you want to see three machine instructions,
10081 including any operands. For convenience, especially when used with
10082 the @code{display} command, the @samp{i} format also prints branch delay
10083 slot instructions, if any, beyond the count specified, which immediately
10084 follow the last instruction that is within the count. The command
10085 @code{disassemble} gives an alternative way of inspecting machine
10086 instructions; see @ref{Machine Code,,Source and Machine Code}.
10088 If a negative repeat count is specified for the formats @samp{s} or @samp{i},
10089 the command displays null-terminated strings or instructions before the given
10090 address as many as the absolute value of the given number. For the @samp{i}
10091 format, we use line number information in the debug info to accurately locate
10092 instruction boundaries while disassembling backward. If line info is not
10093 available, the command stops examining memory with an error message.
10095 All the defaults for the arguments to @code{x} are designed to make it
10096 easy to continue scanning memory with minimal specifications each time
10097 you use @code{x}. For example, after you have inspected three machine
10098 instructions with @samp{x/3i @var{addr}}, you can inspect the next seven
10099 with just @samp{x/7}. If you use @key{RET} to repeat the @code{x} command,
10100 the repeat count @var{n} is used again; the other arguments default as
10101 for successive uses of @code{x}.
10103 When examining machine instructions, the instruction at current program
10104 counter is shown with a @code{=>} marker. For example:
10107 (@value{GDBP}) x/5i $pc-6
10108 0x804837f <main+11>: mov %esp,%ebp
10109 0x8048381 <main+13>: push %ecx
10110 0x8048382 <main+14>: sub $0x4,%esp
10111 => 0x8048385 <main+17>: movl $0x8048460,(%esp)
10112 0x804838c <main+24>: call 0x80482d4 <puts@@plt>
10115 @cindex @code{$_}, @code{$__}, and value history
10116 The addresses and contents printed by the @code{x} command are not saved
10117 in the value history because there is often too much of them and they
10118 would get in the way. Instead, @value{GDBN} makes these values available for
10119 subsequent use in expressions as values of the convenience variables
10120 @code{$_} and @code{$__}. After an @code{x} command, the last address
10121 examined is available for use in expressions in the convenience variable
10122 @code{$_}. The contents of that address, as examined, are available in
10123 the convenience variable @code{$__}.
10125 If the @code{x} command has a repeat count, the address and contents saved
10126 are from the last memory unit printed; this is not the same as the last
10127 address printed if several units were printed on the last line of output.
10129 @anchor{addressable memory unit}
10130 @cindex addressable memory unit
10131 Most targets have an addressable memory unit size of 8 bits. This means
10132 that to each memory address are associated 8 bits of data. Some
10133 targets, however, have other addressable memory unit sizes.
10134 Within @value{GDBN} and this document, the term
10135 @dfn{addressable memory unit} (or @dfn{memory unit} for short) is used
10136 when explicitly referring to a chunk of data of that size. The word
10137 @dfn{byte} is used to refer to a chunk of data of 8 bits, regardless of
10138 the addressable memory unit size of the target. For most systems,
10139 addressable memory unit is a synonym of byte.
10141 @cindex remote memory comparison
10142 @cindex target memory comparison
10143 @cindex verify remote memory image
10144 @cindex verify target memory image
10145 When you are debugging a program running on a remote target machine
10146 (@pxref{Remote Debugging}), you may wish to verify the program's image
10147 in the remote machine's memory against the executable file you
10148 downloaded to the target. Or, on any target, you may want to check
10149 whether the program has corrupted its own read-only sections. The
10150 @code{compare-sections} command is provided for such situations.
10153 @kindex compare-sections
10154 @item compare-sections @r{[}@var{section-name}@r{|}@code{-r}@r{]}
10155 Compare the data of a loadable section @var{section-name} in the
10156 executable file of the program being debugged with the same section in
10157 the target machine's memory, and report any mismatches. With no
10158 arguments, compares all loadable sections. With an argument of
10159 @code{-r}, compares all loadable read-only sections.
10161 Note: for remote targets, this command can be accelerated if the
10162 target supports computing the CRC checksum of a block of memory
10163 (@pxref{qCRC packet}).
10167 @section Automatic Display
10168 @cindex automatic display
10169 @cindex display of expressions
10171 If you find that you want to print the value of an expression frequently
10172 (to see how it changes), you might want to add it to the @dfn{automatic
10173 display list} so that @value{GDBN} prints its value each time your program stops.
10174 Each expression added to the list is given a number to identify it;
10175 to remove an expression from the list, you specify that number.
10176 The automatic display looks like this:
10180 3: bar[5] = (struct hack *) 0x3804
10184 This display shows item numbers, expressions and their current values. As with
10185 displays you request manually using @code{x} or @code{print}, you can
10186 specify the output format you prefer; in fact, @code{display} decides
10187 whether to use @code{print} or @code{x} depending your format
10188 specification---it uses @code{x} if you specify either the @samp{i}
10189 or @samp{s} format, or a unit size; otherwise it uses @code{print}.
10193 @item display @var{expr}
10194 Add the expression @var{expr} to the list of expressions to display
10195 each time your program stops. @xref{Expressions, ,Expressions}.
10197 @code{display} does not repeat if you press @key{RET} again after using it.
10199 @item display/@var{fmt} @var{expr}
10200 For @var{fmt} specifying only a display format and not a size or
10201 count, add the expression @var{expr} to the auto-display list but
10202 arrange to display it each time in the specified format @var{fmt}.
10203 @xref{Output Formats,,Output Formats}.
10205 @item display/@var{fmt} @var{addr}
10206 For @var{fmt} @samp{i} or @samp{s}, or including a unit-size or a
10207 number of units, add the expression @var{addr} as a memory address to
10208 be examined each time your program stops. Examining means in effect
10209 doing @samp{x/@var{fmt} @var{addr}}. @xref{Memory, ,Examining Memory}.
10212 For example, @samp{display/i $pc} can be helpful, to see the machine
10213 instruction about to be executed each time execution stops (@samp{$pc}
10214 is a common name for the program counter; @pxref{Registers, ,Registers}).
10217 @kindex delete display
10219 @item undisplay @var{dnums}@dots{}
10220 @itemx delete display @var{dnums}@dots{}
10221 Remove items from the list of expressions to display. Specify the
10222 numbers of the displays that you want affected with the command
10223 argument @var{dnums}. It can be a single display number, one of the
10224 numbers shown in the first field of the @samp{info display} display;
10225 or it could be a range of display numbers, as in @code{2-4}.
10227 @code{undisplay} does not repeat if you press @key{RET} after using it.
10228 (Otherwise you would just get the error @samp{No display number @dots{}}.)
10230 @kindex disable display
10231 @item disable display @var{dnums}@dots{}
10232 Disable the display of item numbers @var{dnums}. A disabled display
10233 item is not printed automatically, but is not forgotten. It may be
10234 enabled again later. Specify the numbers of the displays that you
10235 want affected with the command argument @var{dnums}. It can be a
10236 single display number, one of the numbers shown in the first field of
10237 the @samp{info display} display; or it could be a range of display
10238 numbers, as in @code{2-4}.
10240 @kindex enable display
10241 @item enable display @var{dnums}@dots{}
10242 Enable display of item numbers @var{dnums}. It becomes effective once
10243 again in auto display of its expression, until you specify otherwise.
10244 Specify the numbers of the displays that you want affected with the
10245 command argument @var{dnums}. It can be a single display number, one
10246 of the numbers shown in the first field of the @samp{info display}
10247 display; or it could be a range of display numbers, as in @code{2-4}.
10250 Display the current values of the expressions on the list, just as is
10251 done when your program stops.
10253 @kindex info display
10255 Print the list of expressions previously set up to display
10256 automatically, each one with its item number, but without showing the
10257 values. This includes disabled expressions, which are marked as such.
10258 It also includes expressions which would not be displayed right now
10259 because they refer to automatic variables not currently available.
10262 @cindex display disabled out of scope
10263 If a display expression refers to local variables, then it does not make
10264 sense outside the lexical context for which it was set up. Such an
10265 expression is disabled when execution enters a context where one of its
10266 variables is not defined. For example, if you give the command
10267 @code{display last_char} while inside a function with an argument
10268 @code{last_char}, @value{GDBN} displays this argument while your program
10269 continues to stop inside that function. When it stops elsewhere---where
10270 there is no variable @code{last_char}---the display is disabled
10271 automatically. The next time your program stops where @code{last_char}
10272 is meaningful, you can enable the display expression once again.
10274 @node Print Settings
10275 @section Print Settings
10277 @cindex format options
10278 @cindex print settings
10279 @value{GDBN} provides the following ways to control how arrays, structures,
10280 and symbols are printed.
10283 These settings are useful for debugging programs in any language:
10287 @item set print address
10288 @itemx set print address on
10289 @cindex print/don't print memory addresses
10290 @value{GDBN} prints memory addresses showing the location of stack
10291 traces, structure values, pointer values, breakpoints, and so forth,
10292 even when it also displays the contents of those addresses. The default
10293 is @code{on}. For example, this is what a stack frame display looks like with
10294 @code{set print address on}:
10299 #0 set_quotes (lq=0x34c78 "<<", rq=0x34c88 ">>")
10301 530 if (lquote != def_lquote)
10305 @item set print address off
10306 Do not print addresses when displaying their contents. For example,
10307 this is the same stack frame displayed with @code{set print address off}:
10311 (@value{GDBP}) set print addr off
10313 #0 set_quotes (lq="<<", rq=">>") at input.c:530
10314 530 if (lquote != def_lquote)
10318 You can use @samp{set print address off} to eliminate all machine
10319 dependent displays from the @value{GDBN} interface. For example, with
10320 @code{print address off}, you should get the same text for backtraces on
10321 all machines---whether or not they involve pointer arguments.
10324 @item show print address
10325 Show whether or not addresses are to be printed.
10328 When @value{GDBN} prints a symbolic address, it normally prints the
10329 closest earlier symbol plus an offset. If that symbol does not uniquely
10330 identify the address (for example, it is a name whose scope is a single
10331 source file), you may need to clarify. One way to do this is with
10332 @code{info line}, for example @samp{info line *0x4537}. Alternately,
10333 you can set @value{GDBN} to print the source file and line number when
10334 it prints a symbolic address:
10337 @item set print symbol-filename on
10338 @cindex source file and line of a symbol
10339 @cindex symbol, source file and line
10340 Tell @value{GDBN} to print the source file name and line number of a
10341 symbol in the symbolic form of an address.
10343 @item set print symbol-filename off
10344 Do not print source file name and line number of a symbol. This is the
10347 @item show print symbol-filename
10348 Show whether or not @value{GDBN} will print the source file name and
10349 line number of a symbol in the symbolic form of an address.
10352 Another situation where it is helpful to show symbol filenames and line
10353 numbers is when disassembling code; @value{GDBN} shows you the line
10354 number and source file that corresponds to each instruction.
10356 Also, you may wish to see the symbolic form only if the address being
10357 printed is reasonably close to the closest earlier symbol:
10360 @item set print max-symbolic-offset @var{max-offset}
10361 @itemx set print max-symbolic-offset unlimited
10362 @cindex maximum value for offset of closest symbol
10363 Tell @value{GDBN} to only display the symbolic form of an address if the
10364 offset between the closest earlier symbol and the address is less than
10365 @var{max-offset}. The default is @code{unlimited}, which tells @value{GDBN}
10366 to always print the symbolic form of an address if any symbol precedes
10367 it. Zero is equivalent to @code{unlimited}.
10369 @item show print max-symbolic-offset
10370 Ask how large the maximum offset is that @value{GDBN} prints in a
10374 @cindex wild pointer, interpreting
10375 @cindex pointer, finding referent
10376 If you have a pointer and you are not sure where it points, try
10377 @samp{set print symbol-filename on}. Then you can determine the name
10378 and source file location of the variable where it points, using
10379 @samp{p/a @var{pointer}}. This interprets the address in symbolic form.
10380 For example, here @value{GDBN} shows that a variable @code{ptt} points
10381 at another variable @code{t}, defined in @file{hi2.c}:
10384 (@value{GDBP}) set print symbol-filename on
10385 (@value{GDBP}) p/a ptt
10386 $4 = 0xe008 <t in hi2.c>
10390 @emph{Warning:} For pointers that point to a local variable, @samp{p/a}
10391 does not show the symbol name and filename of the referent, even with
10392 the appropriate @code{set print} options turned on.
10395 You can also enable @samp{/a}-like formatting all the time using
10396 @samp{set print symbol on}:
10399 @item set print symbol on
10400 Tell @value{GDBN} to print the symbol corresponding to an address, if
10403 @item set print symbol off
10404 Tell @value{GDBN} not to print the symbol corresponding to an
10405 address. In this mode, @value{GDBN} will still print the symbol
10406 corresponding to pointers to functions. This is the default.
10408 @item show print symbol
10409 Show whether @value{GDBN} will display the symbol corresponding to an
10413 Other settings control how different kinds of objects are printed:
10416 @item set print array
10417 @itemx set print array on
10418 @cindex pretty print arrays
10419 Pretty print arrays. This format is more convenient to read,
10420 but uses more space. The default is off.
10422 @item set print array off
10423 Return to compressed format for arrays.
10425 @item show print array
10426 Show whether compressed or pretty format is selected for displaying
10429 @cindex print array indexes
10430 @item set print array-indexes
10431 @itemx set print array-indexes on
10432 Print the index of each element when displaying arrays. May be more
10433 convenient to locate a given element in the array or quickly find the
10434 index of a given element in that printed array. The default is off.
10436 @item set print array-indexes off
10437 Stop printing element indexes when displaying arrays.
10439 @item show print array-indexes
10440 Show whether the index of each element is printed when displaying
10443 @item set print elements @var{number-of-elements}
10444 @itemx set print elements unlimited
10445 @cindex number of array elements to print
10446 @cindex limit on number of printed array elements
10447 Set a limit on how many elements of an array @value{GDBN} will print.
10448 If @value{GDBN} is printing a large array, it stops printing after it has
10449 printed the number of elements set by the @code{set print elements} command.
10450 This limit also applies to the display of strings.
10451 When @value{GDBN} starts, this limit is set to 200.
10452 Setting @var{number-of-elements} to @code{unlimited} or zero means
10453 that the number of elements to print is unlimited.
10455 @item show print elements
10456 Display the number of elements of a large array that @value{GDBN} will print.
10457 If the number is 0, then the printing is unlimited.
10459 @item set print frame-arguments @var{value}
10460 @kindex set print frame-arguments
10461 @cindex printing frame argument values
10462 @cindex print all frame argument values
10463 @cindex print frame argument values for scalars only
10464 @cindex do not print frame argument values
10465 This command allows to control how the values of arguments are printed
10466 when the debugger prints a frame (@pxref{Frames}). The possible
10471 The values of all arguments are printed.
10474 Print the value of an argument only if it is a scalar. The value of more
10475 complex arguments such as arrays, structures, unions, etc, is replaced
10476 by @code{@dots{}}. This is the default. Here is an example where
10477 only scalar arguments are shown:
10480 #1 0x08048361 in call_me (i=3, s=@dots{}, ss=0xbf8d508c, u=@dots{}, e=green)
10485 None of the argument values are printed. Instead, the value of each argument
10486 is replaced by @code{@dots{}}. In this case, the example above now becomes:
10489 #1 0x08048361 in call_me (i=@dots{}, s=@dots{}, ss=@dots{}, u=@dots{}, e=@dots{})
10494 By default, only scalar arguments are printed. This command can be used
10495 to configure the debugger to print the value of all arguments, regardless
10496 of their type. However, it is often advantageous to not print the value
10497 of more complex parameters. For instance, it reduces the amount of
10498 information printed in each frame, making the backtrace more readable.
10499 Also, it improves performance when displaying Ada frames, because
10500 the computation of large arguments can sometimes be CPU-intensive,
10501 especially in large applications. Setting @code{print frame-arguments}
10502 to @code{scalars} (the default) or @code{none} avoids this computation,
10503 thus speeding up the display of each Ada frame.
10505 @item show print frame-arguments
10506 Show how the value of arguments should be displayed when printing a frame.
10508 @item set print raw frame-arguments on
10509 Print frame arguments in raw, non pretty-printed, form.
10511 @item set print raw frame-arguments off
10512 Print frame arguments in pretty-printed form, if there is a pretty-printer
10513 for the value (@pxref{Pretty Printing}),
10514 otherwise print the value in raw form.
10515 This is the default.
10517 @item show print raw frame-arguments
10518 Show whether to print frame arguments in raw form.
10520 @anchor{set print entry-values}
10521 @item set print entry-values @var{value}
10522 @kindex set print entry-values
10523 Set printing of frame argument values at function entry. In some cases
10524 @value{GDBN} can determine the value of function argument which was passed by
10525 the function caller, even if the value was modified inside the called function
10526 and therefore is different. With optimized code, the current value could be
10527 unavailable, but the entry value may still be known.
10529 The default value is @code{default} (see below for its description). Older
10530 @value{GDBN} behaved as with the setting @code{no}. Compilers not supporting
10531 this feature will behave in the @code{default} setting the same way as with the
10534 This functionality is currently supported only by DWARF 2 debugging format and
10535 the compiler has to produce @samp{DW_TAG_call_site} tags. With
10536 @value{NGCC}, you need to specify @option{-O -g} during compilation, to get
10539 The @var{value} parameter can be one of the following:
10543 Print only actual parameter values, never print values from function entry
10547 #0 different (val=6)
10548 #0 lost (val=<optimized out>)
10550 #0 invalid (val=<optimized out>)
10554 Print only parameter values from function entry point. The actual parameter
10555 values are never printed.
10557 #0 equal (val@@entry=5)
10558 #0 different (val@@entry=5)
10559 #0 lost (val@@entry=5)
10560 #0 born (val@@entry=<optimized out>)
10561 #0 invalid (val@@entry=<optimized out>)
10565 Print only parameter values from function entry point. If value from function
10566 entry point is not known while the actual value is known, print the actual
10567 value for such parameter.
10569 #0 equal (val@@entry=5)
10570 #0 different (val@@entry=5)
10571 #0 lost (val@@entry=5)
10573 #0 invalid (val@@entry=<optimized out>)
10577 Print actual parameter values. If actual parameter value is not known while
10578 value from function entry point is known, print the entry point value for such
10582 #0 different (val=6)
10583 #0 lost (val@@entry=5)
10585 #0 invalid (val=<optimized out>)
10589 Always print both the actual parameter value and its value from function entry
10590 point, even if values of one or both are not available due to compiler
10593 #0 equal (val=5, val@@entry=5)
10594 #0 different (val=6, val@@entry=5)
10595 #0 lost (val=<optimized out>, val@@entry=5)
10596 #0 born (val=10, val@@entry=<optimized out>)
10597 #0 invalid (val=<optimized out>, val@@entry=<optimized out>)
10601 Print the actual parameter value if it is known and also its value from
10602 function entry point if it is known. If neither is known, print for the actual
10603 value @code{<optimized out>}. If not in MI mode (@pxref{GDB/MI}) and if both
10604 values are known and identical, print the shortened
10605 @code{param=param@@entry=VALUE} notation.
10607 #0 equal (val=val@@entry=5)
10608 #0 different (val=6, val@@entry=5)
10609 #0 lost (val@@entry=5)
10611 #0 invalid (val=<optimized out>)
10615 Always print the actual parameter value. Print also its value from function
10616 entry point, but only if it is known. If not in MI mode (@pxref{GDB/MI}) and
10617 if both values are known and identical, print the shortened
10618 @code{param=param@@entry=VALUE} notation.
10620 #0 equal (val=val@@entry=5)
10621 #0 different (val=6, val@@entry=5)
10622 #0 lost (val=<optimized out>, val@@entry=5)
10624 #0 invalid (val=<optimized out>)
10628 For analysis messages on possible failures of frame argument values at function
10629 entry resolution see @ref{set debug entry-values}.
10631 @item show print entry-values
10632 Show the method being used for printing of frame argument values at function
10635 @item set print repeats @var{number-of-repeats}
10636 @itemx set print repeats unlimited
10637 @cindex repeated array elements
10638 Set the threshold for suppressing display of repeated array
10639 elements. When the number of consecutive identical elements of an
10640 array exceeds the threshold, @value{GDBN} prints the string
10641 @code{"<repeats @var{n} times>"}, where @var{n} is the number of
10642 identical repetitions, instead of displaying the identical elements
10643 themselves. Setting the threshold to @code{unlimited} or zero will
10644 cause all elements to be individually printed. The default threshold
10647 @item show print repeats
10648 Display the current threshold for printing repeated identical
10651 @item set print max-depth @var{depth}
10652 @item set print max-depth unlimited
10653 @cindex printing nested structures
10654 Set the threshold after which nested structures are replaced with
10655 ellipsis, this can make visualising deeply nested structures easier.
10657 For example, given this C code
10660 typedef struct s1 @{ int a; @} s1;
10661 typedef struct s2 @{ s1 b; @} s2;
10662 typedef struct s3 @{ s2 c; @} s3;
10663 typedef struct s4 @{ s3 d; @} s4;
10665 s4 var = @{ @{ @{ @{ 3 @} @} @} @};
10668 The following table shows how different values of @var{depth} will
10669 effect how @code{var} is printed by @value{GDBN}:
10671 @multitable @columnfractions .3 .7
10672 @headitem @var{depth} setting @tab Result of @samp{p var}
10674 @tab @code{$1 = @{d = @{c = @{b = @{a = 3@}@}@}@}}
10676 @tab @code{$1 = @{...@}}
10678 @tab @code{$1 = @{d = @{...@}@}}
10680 @tab @code{$1 = @{d = @{c = @{...@}@}@}}
10682 @tab @code{$1 = @{d = @{c = @{b = @{...@}@}@}@}}
10684 @tab @code{$1 = @{d = @{c = @{b = @{a = 3@}@}@}@}}
10687 To see the contents of structures that have been hidden the user can
10688 either increase the print max-depth, or they can print the elements of
10689 the structure that are visible, for example
10692 (gdb) set print max-depth 2
10694 $1 = @{d = @{c = @{...@}@}@}
10696 $2 = @{c = @{b = @{...@}@}@}
10698 $3 = @{b = @{a = 3@}@}
10701 The pattern used to replace nested structures varies based on
10702 language, for most languages @code{@{...@}} is used, but Fortran uses
10705 @item show print max-depth
10706 Display the current threshold after which nested structures are
10707 replaces with ellipsis.
10709 @item set print null-stop
10710 @cindex @sc{null} elements in arrays
10711 Cause @value{GDBN} to stop printing the characters of an array when the first
10712 @sc{null} is encountered. This is useful when large arrays actually
10713 contain only short strings.
10714 The default is off.
10716 @item show print null-stop
10717 Show whether @value{GDBN} stops printing an array on the first
10718 @sc{null} character.
10720 @item set print pretty on
10721 @cindex print structures in indented form
10722 @cindex indentation in structure display
10723 Cause @value{GDBN} to print structures in an indented format with one member
10724 per line, like this:
10739 @item set print pretty off
10740 Cause @value{GDBN} to print structures in a compact format, like this:
10744 $1 = @{next = 0x0, flags = @{sweet = 1, sour = 1@}, \
10745 meat = 0x54 "Pork"@}
10750 This is the default format.
10752 @item show print pretty
10753 Show which format @value{GDBN} is using to print structures.
10755 @item set print sevenbit-strings on
10756 @cindex eight-bit characters in strings
10757 @cindex octal escapes in strings
10758 Print using only seven-bit characters; if this option is set,
10759 @value{GDBN} displays any eight-bit characters (in strings or
10760 character values) using the notation @code{\}@var{nnn}. This setting is
10761 best if you are working in English (@sc{ascii}) and you use the
10762 high-order bit of characters as a marker or ``meta'' bit.
10764 @item set print sevenbit-strings off
10765 Print full eight-bit characters. This allows the use of more
10766 international character sets, and is the default.
10768 @item show print sevenbit-strings
10769 Show whether or not @value{GDBN} is printing only seven-bit characters.
10771 @item set print union on
10772 @cindex unions in structures, printing
10773 Tell @value{GDBN} to print unions which are contained in structures
10774 and other unions. This is the default setting.
10776 @item set print union off
10777 Tell @value{GDBN} not to print unions which are contained in
10778 structures and other unions. @value{GDBN} will print @code{"@{...@}"}
10781 @item show print union
10782 Ask @value{GDBN} whether or not it will print unions which are contained in
10783 structures and other unions.
10785 For example, given the declarations
10788 typedef enum @{Tree, Bug@} Species;
10789 typedef enum @{Big_tree, Acorn, Seedling@} Tree_forms;
10790 typedef enum @{Caterpillar, Cocoon, Butterfly@}
10801 struct thing foo = @{Tree, @{Acorn@}@};
10805 with @code{set print union on} in effect @samp{p foo} would print
10808 $1 = @{it = Tree, form = @{tree = Acorn, bug = Cocoon@}@}
10812 and with @code{set print union off} in effect it would print
10815 $1 = @{it = Tree, form = @{...@}@}
10819 @code{set print union} affects programs written in C-like languages
10825 These settings are of interest when debugging C@t{++} programs:
10828 @cindex demangling C@t{++} names
10829 @item set print demangle
10830 @itemx set print demangle on
10831 Print C@t{++} names in their source form rather than in the encoded
10832 (``mangled'') form passed to the assembler and linker for type-safe
10833 linkage. The default is on.
10835 @item show print demangle
10836 Show whether C@t{++} names are printed in mangled or demangled form.
10838 @item set print asm-demangle
10839 @itemx set print asm-demangle on
10840 Print C@t{++} names in their source form rather than their mangled form, even
10841 in assembler code printouts such as instruction disassemblies.
10842 The default is off.
10844 @item show print asm-demangle
10845 Show whether C@t{++} names in assembly listings are printed in mangled
10848 @cindex C@t{++} symbol decoding style
10849 @cindex symbol decoding style, C@t{++}
10850 @kindex set demangle-style
10851 @item set demangle-style @var{style}
10852 Choose among several encoding schemes used by different compilers to represent
10853 C@t{++} names. If you omit @var{style}, you will see a list of possible
10854 formats. The default value is @var{auto}, which lets @value{GDBN} choose a
10855 decoding style by inspecting your program.
10857 @item show demangle-style
10858 Display the encoding style currently in use for decoding C@t{++} symbols.
10860 @item set print object
10861 @itemx set print object on
10862 @cindex derived type of an object, printing
10863 @cindex display derived types
10864 When displaying a pointer to an object, identify the @emph{actual}
10865 (derived) type of the object rather than the @emph{declared} type, using
10866 the virtual function table. Note that the virtual function table is
10867 required---this feature can only work for objects that have run-time
10868 type identification; a single virtual method in the object's declared
10869 type is sufficient. Note that this setting is also taken into account when
10870 working with variable objects via MI (@pxref{GDB/MI}).
10872 @item set print object off
10873 Display only the declared type of objects, without reference to the
10874 virtual function table. This is the default setting.
10876 @item show print object
10877 Show whether actual, or declared, object types are displayed.
10879 @item set print static-members
10880 @itemx set print static-members on
10881 @cindex static members of C@t{++} objects
10882 Print static members when displaying a C@t{++} object. The default is on.
10884 @item set print static-members off
10885 Do not print static members when displaying a C@t{++} object.
10887 @item show print static-members
10888 Show whether C@t{++} static members are printed or not.
10890 @item set print pascal_static-members
10891 @itemx set print pascal_static-members on
10892 @cindex static members of Pascal objects
10893 @cindex Pascal objects, static members display
10894 Print static members when displaying a Pascal object. The default is on.
10896 @item set print pascal_static-members off
10897 Do not print static members when displaying a Pascal object.
10899 @item show print pascal_static-members
10900 Show whether Pascal static members are printed or not.
10902 @c These don't work with HP ANSI C++ yet.
10903 @item set print vtbl
10904 @itemx set print vtbl on
10905 @cindex pretty print C@t{++} virtual function tables
10906 @cindex virtual functions (C@t{++}) display
10907 @cindex VTBL display
10908 Pretty print C@t{++} virtual function tables. The default is off.
10909 (The @code{vtbl} commands do not work on programs compiled with the HP
10910 ANSI C@t{++} compiler (@code{aCC}).)
10912 @item set print vtbl off
10913 Do not pretty print C@t{++} virtual function tables.
10915 @item show print vtbl
10916 Show whether C@t{++} virtual function tables are pretty printed, or not.
10919 @node Pretty Printing
10920 @section Pretty Printing
10922 @value{GDBN} provides a mechanism to allow pretty-printing of values using
10923 Python code. It greatly simplifies the display of complex objects. This
10924 mechanism works for both MI and the CLI.
10927 * Pretty-Printer Introduction:: Introduction to pretty-printers
10928 * Pretty-Printer Example:: An example pretty-printer
10929 * Pretty-Printer Commands:: Pretty-printer commands
10932 @node Pretty-Printer Introduction
10933 @subsection Pretty-Printer Introduction
10935 When @value{GDBN} prints a value, it first sees if there is a pretty-printer
10936 registered for the value. If there is then @value{GDBN} invokes the
10937 pretty-printer to print the value. Otherwise the value is printed normally.
10939 Pretty-printers are normally named. This makes them easy to manage.
10940 The @samp{info pretty-printer} command will list all the installed
10941 pretty-printers with their names.
10942 If a pretty-printer can handle multiple data types, then its
10943 @dfn{subprinters} are the printers for the individual data types.
10944 Each such subprinter has its own name.
10945 The format of the name is @var{printer-name};@var{subprinter-name}.
10947 Pretty-printers are installed by @dfn{registering} them with @value{GDBN}.
10948 Typically they are automatically loaded and registered when the corresponding
10949 debug information is loaded, thus making them available without having to
10950 do anything special.
10952 There are three places where a pretty-printer can be registered.
10956 Pretty-printers registered globally are available when debugging
10960 Pretty-printers registered with a program space are available only
10961 when debugging that program.
10962 @xref{Progspaces In Python}, for more details on program spaces in Python.
10965 Pretty-printers registered with an objfile are loaded and unloaded
10966 with the corresponding objfile (e.g., shared library).
10967 @xref{Objfiles In Python}, for more details on objfiles in Python.
10970 @xref{Selecting Pretty-Printers}, for further information on how
10971 pretty-printers are selected,
10973 @xref{Writing a Pretty-Printer}, for implementing pretty printers
10976 @node Pretty-Printer Example
10977 @subsection Pretty-Printer Example
10979 Here is how a C@t{++} @code{std::string} looks without a pretty-printer:
10982 (@value{GDBP}) print s
10984 static npos = 4294967295,
10986 <std::allocator<char>> = @{
10987 <__gnu_cxx::new_allocator<char>> = @{
10988 <No data fields>@}, <No data fields>
10990 members of std::basic_string<char, std::char_traits<char>,
10991 std::allocator<char> >::_Alloc_hider:
10992 _M_p = 0x804a014 "abcd"
10997 With a pretty-printer for @code{std::string} only the contents are printed:
11000 (@value{GDBP}) print s
11004 @node Pretty-Printer Commands
11005 @subsection Pretty-Printer Commands
11006 @cindex pretty-printer commands
11009 @kindex info pretty-printer
11010 @item info pretty-printer [@var{object-regexp} [@var{name-regexp}]]
11011 Print the list of installed pretty-printers.
11012 This includes disabled pretty-printers, which are marked as such.
11014 @var{object-regexp} is a regular expression matching the objects
11015 whose pretty-printers to list.
11016 Objects can be @code{global}, the program space's file
11017 (@pxref{Progspaces In Python}),
11018 and the object files within that program space (@pxref{Objfiles In Python}).
11019 @xref{Selecting Pretty-Printers}, for details on how @value{GDBN}
11020 looks up a printer from these three objects.
11022 @var{name-regexp} is a regular expression matching the name of the printers
11025 @kindex disable pretty-printer
11026 @item disable pretty-printer [@var{object-regexp} [@var{name-regexp}]]
11027 Disable pretty-printers matching @var{object-regexp} and @var{name-regexp}.
11028 A disabled pretty-printer is not forgotten, it may be enabled again later.
11030 @kindex enable pretty-printer
11031 @item enable pretty-printer [@var{object-regexp} [@var{name-regexp}]]
11032 Enable pretty-printers matching @var{object-regexp} and @var{name-regexp}.
11037 Suppose we have three pretty-printers installed: one from library1.so
11038 named @code{foo} that prints objects of type @code{foo}, and
11039 another from library2.so named @code{bar} that prints two types of objects,
11040 @code{bar1} and @code{bar2}.
11043 (gdb) info pretty-printer
11050 (gdb) info pretty-printer library2
11055 (gdb) disable pretty-printer library1
11057 2 of 3 printers enabled
11058 (gdb) info pretty-printer
11065 (gdb) disable pretty-printer library2 bar;bar1
11067 1 of 3 printers enabled
11068 (gdb) info pretty-printer library2
11075 (gdb) disable pretty-printer library2 bar
11077 0 of 3 printers enabled
11078 (gdb) info pretty-printer library2
11087 Note that for @code{bar} the entire printer can be disabled,
11088 as can each individual subprinter.
11090 @node Value History
11091 @section Value History
11093 @cindex value history
11094 @cindex history of values printed by @value{GDBN}
11095 Values printed by the @code{print} command are saved in the @value{GDBN}
11096 @dfn{value history}. This allows you to refer to them in other expressions.
11097 Values are kept until the symbol table is re-read or discarded
11098 (for example with the @code{file} or @code{symbol-file} commands).
11099 When the symbol table changes, the value history is discarded,
11100 since the values may contain pointers back to the types defined in the
11105 @cindex history number
11106 The values printed are given @dfn{history numbers} by which you can
11107 refer to them. These are successive integers starting with one.
11108 @code{print} shows you the history number assigned to a value by
11109 printing @samp{$@var{num} = } before the value; here @var{num} is the
11112 To refer to any previous value, use @samp{$} followed by the value's
11113 history number. The way @code{print} labels its output is designed to
11114 remind you of this. Just @code{$} refers to the most recent value in
11115 the history, and @code{$$} refers to the value before that.
11116 @code{$$@var{n}} refers to the @var{n}th value from the end; @code{$$2}
11117 is the value just prior to @code{$$}, @code{$$1} is equivalent to
11118 @code{$$}, and @code{$$0} is equivalent to @code{$}.
11120 For example, suppose you have just printed a pointer to a structure and
11121 want to see the contents of the structure. It suffices to type
11127 If you have a chain of structures where the component @code{next} points
11128 to the next one, you can print the contents of the next one with this:
11135 You can print successive links in the chain by repeating this
11136 command---which you can do by just typing @key{RET}.
11138 Note that the history records values, not expressions. If the value of
11139 @code{x} is 4 and you type these commands:
11147 then the value recorded in the value history by the @code{print} command
11148 remains 4 even though the value of @code{x} has changed.
11151 @kindex show values
11153 Print the last ten values in the value history, with their item numbers.
11154 This is like @samp{p@ $$9} repeated ten times, except that @code{show
11155 values} does not change the history.
11157 @item show values @var{n}
11158 Print ten history values centered on history item number @var{n}.
11160 @item show values +
11161 Print ten history values just after the values last printed. If no more
11162 values are available, @code{show values +} produces no display.
11165 Pressing @key{RET} to repeat @code{show values @var{n}} has exactly the
11166 same effect as @samp{show values +}.
11168 @node Convenience Vars
11169 @section Convenience Variables
11171 @cindex convenience variables
11172 @cindex user-defined variables
11173 @value{GDBN} provides @dfn{convenience variables} that you can use within
11174 @value{GDBN} to hold on to a value and refer to it later. These variables
11175 exist entirely within @value{GDBN}; they are not part of your program, and
11176 setting a convenience variable has no direct effect on further execution
11177 of your program. That is why you can use them freely.
11179 Convenience variables are prefixed with @samp{$}. Any name preceded by
11180 @samp{$} can be used for a convenience variable, unless it is one of
11181 the predefined machine-specific register names (@pxref{Registers, ,Registers}).
11182 (Value history references, in contrast, are @emph{numbers} preceded
11183 by @samp{$}. @xref{Value History, ,Value History}.)
11185 You can save a value in a convenience variable with an assignment
11186 expression, just as you would set a variable in your program.
11190 set $foo = *object_ptr
11194 would save in @code{$foo} the value contained in the object pointed to by
11197 Using a convenience variable for the first time creates it, but its
11198 value is @code{void} until you assign a new value. You can alter the
11199 value with another assignment at any time.
11201 Convenience variables have no fixed types. You can assign a convenience
11202 variable any type of value, including structures and arrays, even if
11203 that variable already has a value of a different type. The convenience
11204 variable, when used as an expression, has the type of its current value.
11207 @kindex show convenience
11208 @cindex show all user variables and functions
11209 @item show convenience
11210 Print a list of convenience variables used so far, and their values,
11211 as well as a list of the convenience functions.
11212 Abbreviated @code{show conv}.
11214 @kindex init-if-undefined
11215 @cindex convenience variables, initializing
11216 @item init-if-undefined $@var{variable} = @var{expression}
11217 Set a convenience variable if it has not already been set. This is useful
11218 for user-defined commands that keep some state. It is similar, in concept,
11219 to using local static variables with initializers in C (except that
11220 convenience variables are global). It can also be used to allow users to
11221 override default values used in a command script.
11223 If the variable is already defined then the expression is not evaluated so
11224 any side-effects do not occur.
11227 One of the ways to use a convenience variable is as a counter to be
11228 incremented or a pointer to be advanced. For example, to print
11229 a field from successive elements of an array of structures:
11233 print bar[$i++]->contents
11237 Repeat that command by typing @key{RET}.
11239 Some convenience variables are created automatically by @value{GDBN} and given
11240 values likely to be useful.
11243 @vindex $_@r{, convenience variable}
11245 The variable @code{$_} is automatically set by the @code{x} command to
11246 the last address examined (@pxref{Memory, ,Examining Memory}). Other
11247 commands which provide a default address for @code{x} to examine also
11248 set @code{$_} to that address; these commands include @code{info line}
11249 and @code{info breakpoint}. The type of @code{$_} is @code{void *}
11250 except when set by the @code{x} command, in which case it is a pointer
11251 to the type of @code{$__}.
11253 @vindex $__@r{, convenience variable}
11255 The variable @code{$__} is automatically set by the @code{x} command
11256 to the value found in the last address examined. Its type is chosen
11257 to match the format in which the data was printed.
11260 @vindex $_exitcode@r{, convenience variable}
11261 When the program being debugged terminates normally, @value{GDBN}
11262 automatically sets this variable to the exit code of the program, and
11263 resets @code{$_exitsignal} to @code{void}.
11266 @vindex $_exitsignal@r{, convenience variable}
11267 When the program being debugged dies due to an uncaught signal,
11268 @value{GDBN} automatically sets this variable to that signal's number,
11269 and resets @code{$_exitcode} to @code{void}.
11271 To distinguish between whether the program being debugged has exited
11272 (i.e., @code{$_exitcode} is not @code{void}) or signalled (i.e.,
11273 @code{$_exitsignal} is not @code{void}), the convenience function
11274 @code{$_isvoid} can be used (@pxref{Convenience Funs,, Convenience
11275 Functions}). For example, considering the following source code:
11278 #include <signal.h>
11281 main (int argc, char *argv[])
11288 A valid way of telling whether the program being debugged has exited
11289 or signalled would be:
11292 (@value{GDBP}) define has_exited_or_signalled
11293 Type commands for definition of ``has_exited_or_signalled''.
11294 End with a line saying just ``end''.
11295 >if $_isvoid ($_exitsignal)
11296 >echo The program has exited\n
11298 >echo The program has signalled\n
11304 Program terminated with signal SIGALRM, Alarm clock.
11305 The program no longer exists.
11306 (@value{GDBP}) has_exited_or_signalled
11307 The program has signalled
11310 As can be seen, @value{GDBN} correctly informs that the program being
11311 debugged has signalled, since it calls @code{raise} and raises a
11312 @code{SIGALRM} signal. If the program being debugged had not called
11313 @code{raise}, then @value{GDBN} would report a normal exit:
11316 (@value{GDBP}) has_exited_or_signalled
11317 The program has exited
11321 The variable @code{$_exception} is set to the exception object being
11322 thrown at an exception-related catchpoint. @xref{Set Catchpoints}.
11325 @itemx $_probe_arg0@dots{}$_probe_arg11
11326 Arguments to a static probe. @xref{Static Probe Points}.
11329 @vindex $_sdata@r{, inspect, convenience variable}
11330 The variable @code{$_sdata} contains extra collected static tracepoint
11331 data. @xref{Tracepoint Actions,,Tracepoint Action Lists}. Note that
11332 @code{$_sdata} could be empty, if not inspecting a trace buffer, or
11333 if extra static tracepoint data has not been collected.
11336 @vindex $_siginfo@r{, convenience variable}
11337 The variable @code{$_siginfo} contains extra signal information
11338 (@pxref{extra signal information}). Note that @code{$_siginfo}
11339 could be empty, if the application has not yet received any signals.
11340 For example, it will be empty before you execute the @code{run} command.
11343 @vindex $_tlb@r{, convenience variable}
11344 The variable @code{$_tlb} is automatically set when debugging
11345 applications running on MS-Windows in native mode or connected to
11346 gdbserver that supports the @code{qGetTIBAddr} request.
11347 @xref{General Query Packets}.
11348 This variable contains the address of the thread information block.
11351 The number of the current inferior. @xref{Inferiors and
11352 Programs, ,Debugging Multiple Inferiors and Programs}.
11355 The thread number of the current thread. @xref{thread numbers}.
11358 The global number of the current thread. @xref{global thread numbers}.
11362 @vindex $_gdb_major@r{, convenience variable}
11363 @vindex $_gdb_minor@r{, convenience variable}
11364 The major and minor version numbers of the running @value{GDBN}.
11365 Development snapshots and pretest versions have their minor version
11366 incremented by one; thus, @value{GDBN} pretest 9.11.90 will produce
11367 the value 12 for @code{$_gdb_minor}. These variables allow you to
11368 write scripts that work with different versions of @value{GDBN}
11369 without errors caused by features unavailable in some of those
11372 @item $_shell_exitcode
11373 @itemx $_shell_exitsignal
11374 @vindex $_shell_exitcode@r{, convenience variable}
11375 @vindex $_shell_exitsignal@r{, convenience variable}
11376 @cindex shell command, exit code
11377 @cindex shell command, exit signal
11378 @cindex exit status of shell commands
11379 @value{GDBN} commands such as @code{shell} and @code{|} are launching
11380 shell commands. When a launched command terminates, @value{GDBN}
11381 automatically maintains the variables @code{$_shell_exitcode}
11382 and @code{$_shell_exitsignal} according to the exit status of the last
11383 launched command. These variables are set and used similarly to
11384 the variables @code{$_exitcode} and @code{$_exitsignal}.
11388 @node Convenience Funs
11389 @section Convenience Functions
11391 @cindex convenience functions
11392 @value{GDBN} also supplies some @dfn{convenience functions}. These
11393 have a syntax similar to convenience variables. A convenience
11394 function can be used in an expression just like an ordinary function;
11395 however, a convenience function is implemented internally to
11398 These functions do not require @value{GDBN} to be configured with
11399 @code{Python} support, which means that they are always available.
11403 @item $_isvoid (@var{expr})
11404 @findex $_isvoid@r{, convenience function}
11405 Return one if the expression @var{expr} is @code{void}. Otherwise it
11408 A @code{void} expression is an expression where the type of the result
11409 is @code{void}. For example, you can examine a convenience variable
11410 (see @ref{Convenience Vars,, Convenience Variables}) to check whether
11414 (@value{GDBP}) print $_exitcode
11416 (@value{GDBP}) print $_isvoid ($_exitcode)
11419 Starting program: ./a.out
11420 [Inferior 1 (process 29572) exited normally]
11421 (@value{GDBP}) print $_exitcode
11423 (@value{GDBP}) print $_isvoid ($_exitcode)
11427 In the example above, we used @code{$_isvoid} to check whether
11428 @code{$_exitcode} is @code{void} before and after the execution of the
11429 program being debugged. Before the execution there is no exit code to
11430 be examined, therefore @code{$_exitcode} is @code{void}. After the
11431 execution the program being debugged returned zero, therefore
11432 @code{$_exitcode} is zero, which means that it is not @code{void}
11435 The @code{void} expression can also be a call of a function from the
11436 program being debugged. For example, given the following function:
11445 The result of calling it inside @value{GDBN} is @code{void}:
11448 (@value{GDBP}) print foo ()
11450 (@value{GDBP}) print $_isvoid (foo ())
11452 (@value{GDBP}) set $v = foo ()
11453 (@value{GDBP}) print $v
11455 (@value{GDBP}) print $_isvoid ($v)
11461 These functions require @value{GDBN} to be configured with
11462 @code{Python} support.
11466 @item $_memeq(@var{buf1}, @var{buf2}, @var{length})
11467 @findex $_memeq@r{, convenience function}
11468 Returns one if the @var{length} bytes at the addresses given by
11469 @var{buf1} and @var{buf2} are equal.
11470 Otherwise it returns zero.
11472 @item $_regex(@var{str}, @var{regex})
11473 @findex $_regex@r{, convenience function}
11474 Returns one if the string @var{str} matches the regular expression
11475 @var{regex}. Otherwise it returns zero.
11476 The syntax of the regular expression is that specified by @code{Python}'s
11477 regular expression support.
11479 @item $_streq(@var{str1}, @var{str2})
11480 @findex $_streq@r{, convenience function}
11481 Returns one if the strings @var{str1} and @var{str2} are equal.
11482 Otherwise it returns zero.
11484 @item $_strlen(@var{str})
11485 @findex $_strlen@r{, convenience function}
11486 Returns the length of string @var{str}.
11488 @item $_caller_is(@var{name}@r{[}, @var{number_of_frames}@r{]})
11489 @findex $_caller_is@r{, convenience function}
11490 Returns one if the calling function's name is equal to @var{name}.
11491 Otherwise it returns zero.
11493 If the optional argument @var{number_of_frames} is provided,
11494 it is the number of frames up in the stack to look.
11502 at testsuite/gdb.python/py-caller-is.c:21
11503 #1 0x00000000004005a0 in middle_func ()
11504 at testsuite/gdb.python/py-caller-is.c:27
11505 #2 0x00000000004005ab in top_func ()
11506 at testsuite/gdb.python/py-caller-is.c:33
11507 #3 0x00000000004005b6 in main ()
11508 at testsuite/gdb.python/py-caller-is.c:39
11509 (gdb) print $_caller_is ("middle_func")
11511 (gdb) print $_caller_is ("top_func", 2)
11515 @item $_caller_matches(@var{regexp}@r{[}, @var{number_of_frames}@r{]})
11516 @findex $_caller_matches@r{, convenience function}
11517 Returns one if the calling function's name matches the regular expression
11518 @var{regexp}. Otherwise it returns zero.
11520 If the optional argument @var{number_of_frames} is provided,
11521 it is the number of frames up in the stack to look.
11524 @item $_any_caller_is(@var{name}@r{[}, @var{number_of_frames}@r{]})
11525 @findex $_any_caller_is@r{, convenience function}
11526 Returns one if any calling function's name is equal to @var{name}.
11527 Otherwise it returns zero.
11529 If the optional argument @var{number_of_frames} is provided,
11530 it is the number of frames up in the stack to look.
11533 This function differs from @code{$_caller_is} in that this function
11534 checks all stack frames from the immediate caller to the frame specified
11535 by @var{number_of_frames}, whereas @code{$_caller_is} only checks the
11536 frame specified by @var{number_of_frames}.
11538 @item $_any_caller_matches(@var{regexp}@r{[}, @var{number_of_frames}@r{]})
11539 @findex $_any_caller_matches@r{, convenience function}
11540 Returns one if any calling function's name matches the regular expression
11541 @var{regexp}. Otherwise it returns zero.
11543 If the optional argument @var{number_of_frames} is provided,
11544 it is the number of frames up in the stack to look.
11547 This function differs from @code{$_caller_matches} in that this function
11548 checks all stack frames from the immediate caller to the frame specified
11549 by @var{number_of_frames}, whereas @code{$_caller_matches} only checks the
11550 frame specified by @var{number_of_frames}.
11552 @item $_as_string(@var{value})
11553 @findex $_as_string@r{, convenience function}
11554 Return the string representation of @var{value}.
11556 This function is useful to obtain the textual label (enumerator) of an
11557 enumeration value. For example, assuming the variable @var{node} is of
11558 an enumerated type:
11561 (gdb) printf "Visiting node of type %s\n", $_as_string(node)
11562 Visiting node of type NODE_INTEGER
11565 @item $_cimag(@var{value})
11566 @itemx $_creal(@var{value})
11567 @findex $_cimag@r{, convenience function}
11568 @findex $_creal@r{, convenience function}
11569 Return the imaginary (@code{$_cimag}) or real (@code{$_creal}) part of
11570 the complex number @var{value}.
11572 The type of the imaginary or real part depends on the type of the
11573 complex number, e.g., using @code{$_cimag} on a @code{float complex}
11574 will return an imaginary part of type @code{float}.
11578 @value{GDBN} provides the ability to list and get help on
11579 convenience functions.
11582 @item help function
11583 @kindex help function
11584 @cindex show all convenience functions
11585 Print a list of all convenience functions.
11592 You can refer to machine register contents, in expressions, as variables
11593 with names starting with @samp{$}. The names of registers are different
11594 for each machine; use @code{info registers} to see the names used on
11598 @kindex info registers
11599 @item info registers
11600 Print the names and values of all registers except floating-point
11601 and vector registers (in the selected stack frame).
11603 @kindex info all-registers
11604 @cindex floating point registers
11605 @item info all-registers
11606 Print the names and values of all registers, including floating-point
11607 and vector registers (in the selected stack frame).
11609 @item info registers @var{reggroup} @dots{}
11610 Print the name and value of the registers in each of the specified
11611 @var{reggroup}s. The @var{reggoup} can be any of those returned by
11612 @code{maint print reggroups} (@pxref{Maintenance Commands}).
11614 @item info registers @var{regname} @dots{}
11615 Print the @dfn{relativized} value of each specified register @var{regname}.
11616 As discussed in detail below, register values are normally relative to
11617 the selected stack frame. The @var{regname} may be any register name valid on
11618 the machine you are using, with or without the initial @samp{$}.
11621 @anchor{standard registers}
11622 @cindex stack pointer register
11623 @cindex program counter register
11624 @cindex process status register
11625 @cindex frame pointer register
11626 @cindex standard registers
11627 @value{GDBN} has four ``standard'' register names that are available (in
11628 expressions) on most machines---whenever they do not conflict with an
11629 architecture's canonical mnemonics for registers. The register names
11630 @code{$pc} and @code{$sp} are used for the program counter register and
11631 the stack pointer. @code{$fp} is used for a register that contains a
11632 pointer to the current stack frame, and @code{$ps} is used for a
11633 register that contains the processor status. For example,
11634 you could print the program counter in hex with
11641 or print the instruction to be executed next with
11648 or add four to the stack pointer@footnote{This is a way of removing
11649 one word from the stack, on machines where stacks grow downward in
11650 memory (most machines, nowadays). This assumes that the innermost
11651 stack frame is selected; setting @code{$sp} is not allowed when other
11652 stack frames are selected. To pop entire frames off the stack,
11653 regardless of machine architecture, use @code{return};
11654 see @ref{Returning, ,Returning from a Function}.} with
11660 Whenever possible, these four standard register names are available on
11661 your machine even though the machine has different canonical mnemonics,
11662 so long as there is no conflict. The @code{info registers} command
11663 shows the canonical names. For example, on the SPARC, @code{info
11664 registers} displays the processor status register as @code{$psr} but you
11665 can also refer to it as @code{$ps}; and on x86-based machines @code{$ps}
11666 is an alias for the @sc{eflags} register.
11668 @value{GDBN} always considers the contents of an ordinary register as an
11669 integer when the register is examined in this way. Some machines have
11670 special registers which can hold nothing but floating point; these
11671 registers are considered to have floating point values. There is no way
11672 to refer to the contents of an ordinary register as floating point value
11673 (although you can @emph{print} it as a floating point value with
11674 @samp{print/f $@var{regname}}).
11676 Some registers have distinct ``raw'' and ``virtual'' data formats. This
11677 means that the data format in which the register contents are saved by
11678 the operating system is not the same one that your program normally
11679 sees. For example, the registers of the 68881 floating point
11680 coprocessor are always saved in ``extended'' (raw) format, but all C
11681 programs expect to work with ``double'' (virtual) format. In such
11682 cases, @value{GDBN} normally works with the virtual format only (the format
11683 that makes sense for your program), but the @code{info registers} command
11684 prints the data in both formats.
11686 @cindex SSE registers (x86)
11687 @cindex MMX registers (x86)
11688 Some machines have special registers whose contents can be interpreted
11689 in several different ways. For example, modern x86-based machines
11690 have SSE and MMX registers that can hold several values packed
11691 together in several different formats. @value{GDBN} refers to such
11692 registers in @code{struct} notation:
11695 (@value{GDBP}) print $xmm1
11697 v4_float = @{0, 3.43859137e-038, 1.54142831e-044, 1.821688e-044@},
11698 v2_double = @{9.92129282474342e-303, 2.7585945287983262e-313@},
11699 v16_int8 = "\000\000\000\000\3706;\001\v\000\000\000\r\000\000",
11700 v8_int16 = @{0, 0, 14072, 315, 11, 0, 13, 0@},
11701 v4_int32 = @{0, 20657912, 11, 13@},
11702 v2_int64 = @{88725056443645952, 55834574859@},
11703 uint128 = 0x0000000d0000000b013b36f800000000
11708 To set values of such registers, you need to tell @value{GDBN} which
11709 view of the register you wish to change, as if you were assigning
11710 value to a @code{struct} member:
11713 (@value{GDBP}) set $xmm1.uint128 = 0x000000000000000000000000FFFFFFFF
11716 Normally, register values are relative to the selected stack frame
11717 (@pxref{Selection, ,Selecting a Frame}). This means that you get the
11718 value that the register would contain if all stack frames farther in
11719 were exited and their saved registers restored. In order to see the
11720 true contents of hardware registers, you must select the innermost
11721 frame (with @samp{frame 0}).
11723 @cindex caller-saved registers
11724 @cindex call-clobbered registers
11725 @cindex volatile registers
11726 @cindex <not saved> values
11727 Usually ABIs reserve some registers as not needed to be saved by the
11728 callee (a.k.a.: ``caller-saved'', ``call-clobbered'' or ``volatile''
11729 registers). It may therefore not be possible for @value{GDBN} to know
11730 the value a register had before the call (in other words, in the outer
11731 frame), if the register value has since been changed by the callee.
11732 @value{GDBN} tries to deduce where the inner frame saved
11733 (``callee-saved'') registers, from the debug info, unwind info, or the
11734 machine code generated by your compiler. If some register is not
11735 saved, and @value{GDBN} knows the register is ``caller-saved'' (via
11736 its own knowledge of the ABI, or because the debug/unwind info
11737 explicitly says the register's value is undefined), @value{GDBN}
11738 displays @w{@samp{<not saved>}} as the register's value. With targets
11739 that @value{GDBN} has no knowledge of the register saving convention,
11740 if a register was not saved by the callee, then its value and location
11741 in the outer frame are assumed to be the same of the inner frame.
11742 This is usually harmless, because if the register is call-clobbered,
11743 the caller either does not care what is in the register after the
11744 call, or has code to restore the value that it does care about. Note,
11745 however, that if you change such a register in the outer frame, you
11746 may also be affecting the inner frame. Also, the more ``outer'' the
11747 frame is you're looking at, the more likely a call-clobbered
11748 register's value is to be wrong, in the sense that it doesn't actually
11749 represent the value the register had just before the call.
11751 @node Floating Point Hardware
11752 @section Floating Point Hardware
11753 @cindex floating point
11755 Depending on the configuration, @value{GDBN} may be able to give
11756 you more information about the status of the floating point hardware.
11761 Display hardware-dependent information about the floating
11762 point unit. The exact contents and layout vary depending on the
11763 floating point chip. Currently, @samp{info float} is supported on
11764 the ARM and x86 machines.
11768 @section Vector Unit
11769 @cindex vector unit
11771 Depending on the configuration, @value{GDBN} may be able to give you
11772 more information about the status of the vector unit.
11775 @kindex info vector
11777 Display information about the vector unit. The exact contents and
11778 layout vary depending on the hardware.
11781 @node OS Information
11782 @section Operating System Auxiliary Information
11783 @cindex OS information
11785 @value{GDBN} provides interfaces to useful OS facilities that can help
11786 you debug your program.
11788 @cindex auxiliary vector
11789 @cindex vector, auxiliary
11790 Some operating systems supply an @dfn{auxiliary vector} to programs at
11791 startup. This is akin to the arguments and environment that you
11792 specify for a program, but contains a system-dependent variety of
11793 binary values that tell system libraries important details about the
11794 hardware, operating system, and process. Each value's purpose is
11795 identified by an integer tag; the meanings are well-known but system-specific.
11796 Depending on the configuration and operating system facilities,
11797 @value{GDBN} may be able to show you this information. For remote
11798 targets, this functionality may further depend on the remote stub's
11799 support of the @samp{qXfer:auxv:read} packet, see
11800 @ref{qXfer auxiliary vector read}.
11805 Display the auxiliary vector of the inferior, which can be either a
11806 live process or a core dump file. @value{GDBN} prints each tag value
11807 numerically, and also shows names and text descriptions for recognized
11808 tags. Some values in the vector are numbers, some bit masks, and some
11809 pointers to strings or other data. @value{GDBN} displays each value in the
11810 most appropriate form for a recognized tag, and in hexadecimal for
11811 an unrecognized tag.
11814 On some targets, @value{GDBN} can access operating system-specific
11815 information and show it to you. The types of information available
11816 will differ depending on the type of operating system running on the
11817 target. The mechanism used to fetch the data is described in
11818 @ref{Operating System Information}. For remote targets, this
11819 functionality depends on the remote stub's support of the
11820 @samp{qXfer:osdata:read} packet, see @ref{qXfer osdata read}.
11824 @item info os @var{infotype}
11826 Display OS information of the requested type.
11828 On @sc{gnu}/Linux, the following values of @var{infotype} are valid:
11830 @anchor{linux info os infotypes}
11832 @kindex info os cpus
11834 Display the list of all CPUs/cores. For each CPU/core, @value{GDBN} prints
11835 the available fields from /proc/cpuinfo. For each supported architecture
11836 different fields are available. Two common entries are processor which gives
11837 CPU number and bogomips; a system constant that is calculated during
11838 kernel initialization.
11840 @kindex info os files
11842 Display the list of open file descriptors on the target. For each
11843 file descriptor, @value{GDBN} prints the identifier of the process
11844 owning the descriptor, the command of the owning process, the value
11845 of the descriptor, and the target of the descriptor.
11847 @kindex info os modules
11849 Display the list of all loaded kernel modules on the target. For each
11850 module, @value{GDBN} prints the module name, the size of the module in
11851 bytes, the number of times the module is used, the dependencies of the
11852 module, the status of the module, and the address of the loaded module
11855 @kindex info os msg
11857 Display the list of all System V message queues on the target. For each
11858 message queue, @value{GDBN} prints the message queue key, the message
11859 queue identifier, the access permissions, the current number of bytes
11860 on the queue, the current number of messages on the queue, the processes
11861 that last sent and received a message on the queue, the user and group
11862 of the owner and creator of the message queue, the times at which a
11863 message was last sent and received on the queue, and the time at which
11864 the message queue was last changed.
11866 @kindex info os processes
11868 Display the list of processes on the target. For each process,
11869 @value{GDBN} prints the process identifier, the name of the user, the
11870 command corresponding to the process, and the list of processor cores
11871 that the process is currently running on. (To understand what these
11872 properties mean, for this and the following info types, please consult
11873 the general @sc{gnu}/Linux documentation.)
11875 @kindex info os procgroups
11877 Display the list of process groups on the target. For each process,
11878 @value{GDBN} prints the identifier of the process group that it belongs
11879 to, the command corresponding to the process group leader, the process
11880 identifier, and the command line of the process. The list is sorted
11881 first by the process group identifier, then by the process identifier,
11882 so that processes belonging to the same process group are grouped together
11883 and the process group leader is listed first.
11885 @kindex info os semaphores
11887 Display the list of all System V semaphore sets on the target. For each
11888 semaphore set, @value{GDBN} prints the semaphore set key, the semaphore
11889 set identifier, the access permissions, the number of semaphores in the
11890 set, the user and group of the owner and creator of the semaphore set,
11891 and the times at which the semaphore set was operated upon and changed.
11893 @kindex info os shm
11895 Display the list of all System V shared-memory regions on the target.
11896 For each shared-memory region, @value{GDBN} prints the region key,
11897 the shared-memory identifier, the access permissions, the size of the
11898 region, the process that created the region, the process that last
11899 attached to or detached from the region, the current number of live
11900 attaches to the region, and the times at which the region was last
11901 attached to, detach from, and changed.
11903 @kindex info os sockets
11905 Display the list of Internet-domain sockets on the target. For each
11906 socket, @value{GDBN} prints the address and port of the local and
11907 remote endpoints, the current state of the connection, the creator of
11908 the socket, the IP address family of the socket, and the type of the
11911 @kindex info os threads
11913 Display the list of threads running on the target. For each thread,
11914 @value{GDBN} prints the identifier of the process that the thread
11915 belongs to, the command of the process, the thread identifier, and the
11916 processor core that it is currently running on. The main thread of a
11917 process is not listed.
11921 If @var{infotype} is omitted, then list the possible values for
11922 @var{infotype} and the kind of OS information available for each
11923 @var{infotype}. If the target does not return a list of possible
11924 types, this command will report an error.
11927 @node Memory Region Attributes
11928 @section Memory Region Attributes
11929 @cindex memory region attributes
11931 @dfn{Memory region attributes} allow you to describe special handling
11932 required by regions of your target's memory. @value{GDBN} uses
11933 attributes to determine whether to allow certain types of memory
11934 accesses; whether to use specific width accesses; and whether to cache
11935 target memory. By default the description of memory regions is
11936 fetched from the target (if the current target supports this), but the
11937 user can override the fetched regions.
11939 Defined memory regions can be individually enabled and disabled. When a
11940 memory region is disabled, @value{GDBN} uses the default attributes when
11941 accessing memory in that region. Similarly, if no memory regions have
11942 been defined, @value{GDBN} uses the default attributes when accessing
11945 When a memory region is defined, it is given a number to identify it;
11946 to enable, disable, or remove a memory region, you specify that number.
11950 @item mem @var{lower} @var{upper} @var{attributes}@dots{}
11951 Define a memory region bounded by @var{lower} and @var{upper} with
11952 attributes @var{attributes}@dots{}, and add it to the list of regions
11953 monitored by @value{GDBN}. Note that @var{upper} == 0 is a special
11954 case: it is treated as the target's maximum memory address.
11955 (0xffff on 16 bit targets, 0xffffffff on 32 bit targets, etc.)
11958 Discard any user changes to the memory regions and use target-supplied
11959 regions, if available, or no regions if the target does not support.
11962 @item delete mem @var{nums}@dots{}
11963 Remove memory regions @var{nums}@dots{} from the list of regions
11964 monitored by @value{GDBN}.
11966 @kindex disable mem
11967 @item disable mem @var{nums}@dots{}
11968 Disable monitoring of memory regions @var{nums}@dots{}.
11969 A disabled memory region is not forgotten.
11970 It may be enabled again later.
11973 @item enable mem @var{nums}@dots{}
11974 Enable monitoring of memory regions @var{nums}@dots{}.
11978 Print a table of all defined memory regions, with the following columns
11982 @item Memory Region Number
11983 @item Enabled or Disabled.
11984 Enabled memory regions are marked with @samp{y}.
11985 Disabled memory regions are marked with @samp{n}.
11988 The address defining the inclusive lower bound of the memory region.
11991 The address defining the exclusive upper bound of the memory region.
11994 The list of attributes set for this memory region.
11999 @subsection Attributes
12001 @subsubsection Memory Access Mode
12002 The access mode attributes set whether @value{GDBN} may make read or
12003 write accesses to a memory region.
12005 While these attributes prevent @value{GDBN} from performing invalid
12006 memory accesses, they do nothing to prevent the target system, I/O DMA,
12007 etc.@: from accessing memory.
12011 Memory is read only.
12013 Memory is write only.
12015 Memory is read/write. This is the default.
12018 @subsubsection Memory Access Size
12019 The access size attribute tells @value{GDBN} to use specific sized
12020 accesses in the memory region. Often memory mapped device registers
12021 require specific sized accesses. If no access size attribute is
12022 specified, @value{GDBN} may use accesses of any size.
12026 Use 8 bit memory accesses.
12028 Use 16 bit memory accesses.
12030 Use 32 bit memory accesses.
12032 Use 64 bit memory accesses.
12035 @c @subsubsection Hardware/Software Breakpoints
12036 @c The hardware/software breakpoint attributes set whether @value{GDBN}
12037 @c will use hardware or software breakpoints for the internal breakpoints
12038 @c used by the step, next, finish, until, etc. commands.
12042 @c Always use hardware breakpoints
12043 @c @item swbreak (default)
12046 @subsubsection Data Cache
12047 The data cache attributes set whether @value{GDBN} will cache target
12048 memory. While this generally improves performance by reducing debug
12049 protocol overhead, it can lead to incorrect results because @value{GDBN}
12050 does not know about volatile variables or memory mapped device
12055 Enable @value{GDBN} to cache target memory.
12057 Disable @value{GDBN} from caching target memory. This is the default.
12060 @subsection Memory Access Checking
12061 @value{GDBN} can be instructed to refuse accesses to memory that is
12062 not explicitly described. This can be useful if accessing such
12063 regions has undesired effects for a specific target, or to provide
12064 better error checking. The following commands control this behaviour.
12067 @kindex set mem inaccessible-by-default
12068 @item set mem inaccessible-by-default [on|off]
12069 If @code{on} is specified, make @value{GDBN} treat memory not
12070 explicitly described by the memory ranges as non-existent and refuse accesses
12071 to such memory. The checks are only performed if there's at least one
12072 memory range defined. If @code{off} is specified, make @value{GDBN}
12073 treat the memory not explicitly described by the memory ranges as RAM.
12074 The default value is @code{on}.
12075 @kindex show mem inaccessible-by-default
12076 @item show mem inaccessible-by-default
12077 Show the current handling of accesses to unknown memory.
12081 @c @subsubsection Memory Write Verification
12082 @c The memory write verification attributes set whether @value{GDBN}
12083 @c will re-reads data after each write to verify the write was successful.
12087 @c @item noverify (default)
12090 @node Dump/Restore Files
12091 @section Copy Between Memory and a File
12092 @cindex dump/restore files
12093 @cindex append data to a file
12094 @cindex dump data to a file
12095 @cindex restore data from a file
12097 You can use the commands @code{dump}, @code{append}, and
12098 @code{restore} to copy data between target memory and a file. The
12099 @code{dump} and @code{append} commands write data to a file, and the
12100 @code{restore} command reads data from a file back into the inferior's
12101 memory. Files may be in binary, Motorola S-record, Intel hex,
12102 Tektronix Hex, or Verilog Hex format; however, @value{GDBN} can only
12103 append to binary files, and cannot read from Verilog Hex files.
12108 @item dump @r{[}@var{format}@r{]} memory @var{filename} @var{start_addr} @var{end_addr}
12109 @itemx dump @r{[}@var{format}@r{]} value @var{filename} @var{expr}
12110 Dump the contents of memory from @var{start_addr} to @var{end_addr},
12111 or the value of @var{expr}, to @var{filename} in the given format.
12113 The @var{format} parameter may be any one of:
12120 Motorola S-record format.
12122 Tektronix Hex format.
12124 Verilog Hex format.
12127 @value{GDBN} uses the same definitions of these formats as the
12128 @sc{gnu} binary utilities, like @samp{objdump} and @samp{objcopy}. If
12129 @var{format} is omitted, @value{GDBN} dumps the data in raw binary
12133 @item append @r{[}binary@r{]} memory @var{filename} @var{start_addr} @var{end_addr}
12134 @itemx append @r{[}binary@r{]} value @var{filename} @var{expr}
12135 Append the contents of memory from @var{start_addr} to @var{end_addr},
12136 or the value of @var{expr}, to the file @var{filename}, in raw binary form.
12137 (@value{GDBN} can only append data to files in raw binary form.)
12140 @item restore @var{filename} @r{[}binary@r{]} @var{bias} @var{start} @var{end}
12141 Restore the contents of file @var{filename} into memory. The
12142 @code{restore} command can automatically recognize any known @sc{bfd}
12143 file format, except for raw binary. To restore a raw binary file you
12144 must specify the optional keyword @code{binary} after the filename.
12146 If @var{bias} is non-zero, its value will be added to the addresses
12147 contained in the file. Binary files always start at address zero, so
12148 they will be restored at address @var{bias}. Other bfd files have
12149 a built-in location; they will be restored at offset @var{bias}
12150 from that location.
12152 If @var{start} and/or @var{end} are non-zero, then only data between
12153 file offset @var{start} and file offset @var{end} will be restored.
12154 These offsets are relative to the addresses in the file, before
12155 the @var{bias} argument is applied.
12159 @node Core File Generation
12160 @section How to Produce a Core File from Your Program
12161 @cindex dump core from inferior
12163 A @dfn{core file} or @dfn{core dump} is a file that records the memory
12164 image of a running process and its process status (register values
12165 etc.). Its primary use is post-mortem debugging of a program that
12166 crashed while it ran outside a debugger. A program that crashes
12167 automatically produces a core file, unless this feature is disabled by
12168 the user. @xref{Files}, for information on invoking @value{GDBN} in
12169 the post-mortem debugging mode.
12171 Occasionally, you may wish to produce a core file of the program you
12172 are debugging in order to preserve a snapshot of its state.
12173 @value{GDBN} has a special command for that.
12177 @kindex generate-core-file
12178 @item generate-core-file [@var{file}]
12179 @itemx gcore [@var{file}]
12180 Produce a core dump of the inferior process. The optional argument
12181 @var{file} specifies the file name where to put the core dump. If not
12182 specified, the file name defaults to @file{core.@var{pid}}, where
12183 @var{pid} is the inferior process ID.
12185 Note that this command is implemented only for some systems (as of
12186 this writing, @sc{gnu}/Linux, FreeBSD, Solaris, and S390).
12188 On @sc{gnu}/Linux, this command can take into account the value of the
12189 file @file{/proc/@var{pid}/coredump_filter} when generating the core
12190 dump (@pxref{set use-coredump-filter}), and by default honors the
12191 @code{VM_DONTDUMP} flag for mappings where it is present in the file
12192 @file{/proc/@var{pid}/smaps} (@pxref{set dump-excluded-mappings}).
12194 @kindex set use-coredump-filter
12195 @anchor{set use-coredump-filter}
12196 @item set use-coredump-filter on
12197 @itemx set use-coredump-filter off
12198 Enable or disable the use of the file
12199 @file{/proc/@var{pid}/coredump_filter} when generating core dump
12200 files. This file is used by the Linux kernel to decide what types of
12201 memory mappings will be dumped or ignored when generating a core dump
12202 file. @var{pid} is the process ID of a currently running process.
12204 To make use of this feature, you have to write in the
12205 @file{/proc/@var{pid}/coredump_filter} file a value, in hexadecimal,
12206 which is a bit mask representing the memory mapping types. If a bit
12207 is set in the bit mask, then the memory mappings of the corresponding
12208 types will be dumped; otherwise, they will be ignored. This
12209 configuration is inherited by child processes. For more information
12210 about the bits that can be set in the
12211 @file{/proc/@var{pid}/coredump_filter} file, please refer to the
12212 manpage of @code{core(5)}.
12214 By default, this option is @code{on}. If this option is turned
12215 @code{off}, @value{GDBN} does not read the @file{coredump_filter} file
12216 and instead uses the same default value as the Linux kernel in order
12217 to decide which pages will be dumped in the core dump file. This
12218 value is currently @code{0x33}, which means that bits @code{0}
12219 (anonymous private mappings), @code{1} (anonymous shared mappings),
12220 @code{4} (ELF headers) and @code{5} (private huge pages) are active.
12221 This will cause these memory mappings to be dumped automatically.
12223 @kindex set dump-excluded-mappings
12224 @anchor{set dump-excluded-mappings}
12225 @item set dump-excluded-mappings on
12226 @itemx set dump-excluded-mappings off
12227 If @code{on} is specified, @value{GDBN} will dump memory mappings
12228 marked with the @code{VM_DONTDUMP} flag. This flag is represented in
12229 the file @file{/proc/@var{pid}/smaps} with the acronym @code{dd}.
12231 The default value is @code{off}.
12234 @node Character Sets
12235 @section Character Sets
12236 @cindex character sets
12238 @cindex translating between character sets
12239 @cindex host character set
12240 @cindex target character set
12242 If the program you are debugging uses a different character set to
12243 represent characters and strings than the one @value{GDBN} uses itself,
12244 @value{GDBN} can automatically translate between the character sets for
12245 you. The character set @value{GDBN} uses we call the @dfn{host
12246 character set}; the one the inferior program uses we call the
12247 @dfn{target character set}.
12249 For example, if you are running @value{GDBN} on a @sc{gnu}/Linux system, which
12250 uses the ISO Latin 1 character set, but you are using @value{GDBN}'s
12251 remote protocol (@pxref{Remote Debugging}) to debug a program
12252 running on an IBM mainframe, which uses the @sc{ebcdic} character set,
12253 then the host character set is Latin-1, and the target character set is
12254 @sc{ebcdic}. If you give @value{GDBN} the command @code{set
12255 target-charset EBCDIC-US}, then @value{GDBN} translates between
12256 @sc{ebcdic} and Latin 1 as you print character or string values, or use
12257 character and string literals in expressions.
12259 @value{GDBN} has no way to automatically recognize which character set
12260 the inferior program uses; you must tell it, using the @code{set
12261 target-charset} command, described below.
12263 Here are the commands for controlling @value{GDBN}'s character set
12267 @item set target-charset @var{charset}
12268 @kindex set target-charset
12269 Set the current target character set to @var{charset}. To display the
12270 list of supported target character sets, type
12271 @kbd{@w{set target-charset @key{TAB}@key{TAB}}}.
12273 @item set host-charset @var{charset}
12274 @kindex set host-charset
12275 Set the current host character set to @var{charset}.
12277 By default, @value{GDBN} uses a host character set appropriate to the
12278 system it is running on; you can override that default using the
12279 @code{set host-charset} command. On some systems, @value{GDBN} cannot
12280 automatically determine the appropriate host character set. In this
12281 case, @value{GDBN} uses @samp{UTF-8}.
12283 @value{GDBN} can only use certain character sets as its host character
12284 set. If you type @kbd{@w{set host-charset @key{TAB}@key{TAB}}},
12285 @value{GDBN} will list the host character sets it supports.
12287 @item set charset @var{charset}
12288 @kindex set charset
12289 Set the current host and target character sets to @var{charset}. As
12290 above, if you type @kbd{@w{set charset @key{TAB}@key{TAB}}},
12291 @value{GDBN} will list the names of the character sets that can be used
12292 for both host and target.
12295 @kindex show charset
12296 Show the names of the current host and target character sets.
12298 @item show host-charset
12299 @kindex show host-charset
12300 Show the name of the current host character set.
12302 @item show target-charset
12303 @kindex show target-charset
12304 Show the name of the current target character set.
12306 @item set target-wide-charset @var{charset}
12307 @kindex set target-wide-charset
12308 Set the current target's wide character set to @var{charset}. This is
12309 the character set used by the target's @code{wchar_t} type. To
12310 display the list of supported wide character sets, type
12311 @kbd{@w{set target-wide-charset @key{TAB}@key{TAB}}}.
12313 @item show target-wide-charset
12314 @kindex show target-wide-charset
12315 Show the name of the current target's wide character set.
12318 Here is an example of @value{GDBN}'s character set support in action.
12319 Assume that the following source code has been placed in the file
12320 @file{charset-test.c}:
12326 = @{72, 101, 108, 108, 111, 44, 32, 119,
12327 111, 114, 108, 100, 33, 10, 0@};
12328 char ibm1047_hello[]
12329 = @{200, 133, 147, 147, 150, 107, 64, 166,
12330 150, 153, 147, 132, 90, 37, 0@};
12334 printf ("Hello, world!\n");
12338 In this program, @code{ascii_hello} and @code{ibm1047_hello} are arrays
12339 containing the string @samp{Hello, world!} followed by a newline,
12340 encoded in the @sc{ascii} and @sc{ibm1047} character sets.
12342 We compile the program, and invoke the debugger on it:
12345 $ gcc -g charset-test.c -o charset-test
12346 $ gdb -nw charset-test
12347 GNU gdb 2001-12-19-cvs
12348 Copyright 2001 Free Software Foundation, Inc.
12353 We can use the @code{show charset} command to see what character sets
12354 @value{GDBN} is currently using to interpret and display characters and
12358 (@value{GDBP}) show charset
12359 The current host and target character set is `ISO-8859-1'.
12363 For the sake of printing this manual, let's use @sc{ascii} as our
12364 initial character set:
12366 (@value{GDBP}) set charset ASCII
12367 (@value{GDBP}) show charset
12368 The current host and target character set is `ASCII'.
12372 Let's assume that @sc{ascii} is indeed the correct character set for our
12373 host system --- in other words, let's assume that if @value{GDBN} prints
12374 characters using the @sc{ascii} character set, our terminal will display
12375 them properly. Since our current target character set is also
12376 @sc{ascii}, the contents of @code{ascii_hello} print legibly:
12379 (@value{GDBP}) print ascii_hello
12380 $1 = 0x401698 "Hello, world!\n"
12381 (@value{GDBP}) print ascii_hello[0]
12386 @value{GDBN} uses the target character set for character and string
12387 literals you use in expressions:
12390 (@value{GDBP}) print '+'
12395 The @sc{ascii} character set uses the number 43 to encode the @samp{+}
12398 @value{GDBN} relies on the user to tell it which character set the
12399 target program uses. If we print @code{ibm1047_hello} while our target
12400 character set is still @sc{ascii}, we get jibberish:
12403 (@value{GDBP}) print ibm1047_hello
12404 $4 = 0x4016a8 "\310\205\223\223\226k@@\246\226\231\223\204Z%"
12405 (@value{GDBP}) print ibm1047_hello[0]
12410 If we invoke the @code{set target-charset} followed by @key{TAB}@key{TAB},
12411 @value{GDBN} tells us the character sets it supports:
12414 (@value{GDBP}) set target-charset
12415 ASCII EBCDIC-US IBM1047 ISO-8859-1
12416 (@value{GDBP}) set target-charset
12419 We can select @sc{ibm1047} as our target character set, and examine the
12420 program's strings again. Now the @sc{ascii} string is wrong, but
12421 @value{GDBN} translates the contents of @code{ibm1047_hello} from the
12422 target character set, @sc{ibm1047}, to the host character set,
12423 @sc{ascii}, and they display correctly:
12426 (@value{GDBP}) set target-charset IBM1047
12427 (@value{GDBP}) show charset
12428 The current host character set is `ASCII'.
12429 The current target character set is `IBM1047'.
12430 (@value{GDBP}) print ascii_hello
12431 $6 = 0x401698 "\110\145%%?\054\040\167?\162%\144\041\012"
12432 (@value{GDBP}) print ascii_hello[0]
12434 (@value{GDBP}) print ibm1047_hello
12435 $8 = 0x4016a8 "Hello, world!\n"
12436 (@value{GDBP}) print ibm1047_hello[0]
12441 As above, @value{GDBN} uses the target character set for character and
12442 string literals you use in expressions:
12445 (@value{GDBP}) print '+'
12450 The @sc{ibm1047} character set uses the number 78 to encode the @samp{+}
12453 @node Caching Target Data
12454 @section Caching Data of Targets
12455 @cindex caching data of targets
12457 @value{GDBN} caches data exchanged between the debugger and a target.
12458 Each cache is associated with the address space of the inferior.
12459 @xref{Inferiors and Programs}, about inferior and address space.
12460 Such caching generally improves performance in remote debugging
12461 (@pxref{Remote Debugging}), because it reduces the overhead of the
12462 remote protocol by bundling memory reads and writes into large chunks.
12463 Unfortunately, simply caching everything would lead to incorrect results,
12464 since @value{GDBN} does not necessarily know anything about volatile
12465 values, memory-mapped I/O addresses, etc. Furthermore, in non-stop mode
12466 (@pxref{Non-Stop Mode}) memory can be changed @emph{while} a gdb command
12468 Therefore, by default, @value{GDBN} only caches data
12469 known to be on the stack@footnote{In non-stop mode, it is moderately
12470 rare for a running thread to modify the stack of a stopped thread
12471 in a way that would interfere with a backtrace, and caching of
12472 stack reads provides a significant speed up of remote backtraces.} or
12473 in the code segment.
12474 Other regions of memory can be explicitly marked as
12475 cacheable; @pxref{Memory Region Attributes}.
12478 @kindex set remotecache
12479 @item set remotecache on
12480 @itemx set remotecache off
12481 This option no longer does anything; it exists for compatibility
12484 @kindex show remotecache
12485 @item show remotecache
12486 Show the current state of the obsolete remotecache flag.
12488 @kindex set stack-cache
12489 @item set stack-cache on
12490 @itemx set stack-cache off
12491 Enable or disable caching of stack accesses. When @code{on}, use
12492 caching. By default, this option is @code{on}.
12494 @kindex show stack-cache
12495 @item show stack-cache
12496 Show the current state of data caching for memory accesses.
12498 @kindex set code-cache
12499 @item set code-cache on
12500 @itemx set code-cache off
12501 Enable or disable caching of code segment accesses. When @code{on},
12502 use caching. By default, this option is @code{on}. This improves
12503 performance of disassembly in remote debugging.
12505 @kindex show code-cache
12506 @item show code-cache
12507 Show the current state of target memory cache for code segment
12510 @kindex info dcache
12511 @item info dcache @r{[}line@r{]}
12512 Print the information about the performance of data cache of the
12513 current inferior's address space. The information displayed
12514 includes the dcache width and depth, and for each cache line, its
12515 number, address, and how many times it was referenced. This
12516 command is useful for debugging the data cache operation.
12518 If a line number is specified, the contents of that line will be
12521 @item set dcache size @var{size}
12522 @cindex dcache size
12523 @kindex set dcache size
12524 Set maximum number of entries in dcache (dcache depth above).
12526 @item set dcache line-size @var{line-size}
12527 @cindex dcache line-size
12528 @kindex set dcache line-size
12529 Set number of bytes each dcache entry caches (dcache width above).
12530 Must be a power of 2.
12532 @item show dcache size
12533 @kindex show dcache size
12534 Show maximum number of dcache entries. @xref{Caching Target Data, info dcache}.
12536 @item show dcache line-size
12537 @kindex show dcache line-size
12538 Show default size of dcache lines.
12542 @node Searching Memory
12543 @section Search Memory
12544 @cindex searching memory
12546 Memory can be searched for a particular sequence of bytes with the
12547 @code{find} command.
12551 @item find @r{[}/@var{sn}@r{]} @var{start_addr}, +@var{len}, @var{val1} @r{[}, @var{val2}, @dots{}@r{]}
12552 @itemx find @r{[}/@var{sn}@r{]} @var{start_addr}, @var{end_addr}, @var{val1} @r{[}, @var{val2}, @dots{}@r{]}
12553 Search memory for the sequence of bytes specified by @var{val1}, @var{val2},
12554 etc. The search begins at address @var{start_addr} and continues for either
12555 @var{len} bytes or through to @var{end_addr} inclusive.
12558 @var{s} and @var{n} are optional parameters.
12559 They may be specified in either order, apart or together.
12562 @item @var{s}, search query size
12563 The size of each search query value.
12569 halfwords (two bytes)
12573 giant words (eight bytes)
12576 All values are interpreted in the current language.
12577 This means, for example, that if the current source language is C/C@t{++}
12578 then searching for the string ``hello'' includes the trailing '\0'.
12579 The null terminator can be removed from searching by using casts,
12580 e.g.: @samp{@{char[5]@}"hello"}.
12582 If the value size is not specified, it is taken from the
12583 value's type in the current language.
12584 This is useful when one wants to specify the search
12585 pattern as a mixture of types.
12586 Note that this means, for example, that in the case of C-like languages
12587 a search for an untyped 0x42 will search for @samp{(int) 0x42}
12588 which is typically four bytes.
12590 @item @var{n}, maximum number of finds
12591 The maximum number of matches to print. The default is to print all finds.
12594 You can use strings as search values. Quote them with double-quotes
12596 The string value is copied into the search pattern byte by byte,
12597 regardless of the endianness of the target and the size specification.
12599 The address of each match found is printed as well as a count of the
12600 number of matches found.
12602 The address of the last value found is stored in convenience variable
12604 A count of the number of matches is stored in @samp{$numfound}.
12606 For example, if stopped at the @code{printf} in this function:
12612 static char hello[] = "hello-hello";
12613 static struct @{ char c; short s; int i; @}
12614 __attribute__ ((packed)) mixed
12615 = @{ 'c', 0x1234, 0x87654321 @};
12616 printf ("%s\n", hello);
12621 you get during debugging:
12624 (gdb) find &hello[0], +sizeof(hello), "hello"
12625 0x804956d <hello.1620+6>
12627 (gdb) find &hello[0], +sizeof(hello), 'h', 'e', 'l', 'l', 'o'
12628 0x8049567 <hello.1620>
12629 0x804956d <hello.1620+6>
12631 (gdb) find &hello[0], +sizeof(hello), @{char[5]@}"hello"
12632 0x8049567 <hello.1620>
12633 0x804956d <hello.1620+6>
12635 (gdb) find /b1 &hello[0], +sizeof(hello), 'h', 0x65, 'l'
12636 0x8049567 <hello.1620>
12638 (gdb) find &mixed, +sizeof(mixed), (char) 'c', (short) 0x1234, (int) 0x87654321
12639 0x8049560 <mixed.1625>
12641 (gdb) print $numfound
12644 $2 = (void *) 0x8049560
12648 @section Value Sizes
12650 Whenever @value{GDBN} prints a value memory will be allocated within
12651 @value{GDBN} to hold the contents of the value. It is possible in
12652 some languages with dynamic typing systems, that an invalid program
12653 may indicate a value that is incorrectly large, this in turn may cause
12654 @value{GDBN} to try and allocate an overly large ammount of memory.
12657 @kindex set max-value-size
12658 @item set max-value-size @var{bytes}
12659 @itemx set max-value-size unlimited
12660 Set the maximum size of memory that @value{GDBN} will allocate for the
12661 contents of a value to @var{bytes}, trying to display a value that
12662 requires more memory than that will result in an error.
12664 Setting this variable does not effect values that have already been
12665 allocated within @value{GDBN}, only future allocations.
12667 There's a minimum size that @code{max-value-size} can be set to in
12668 order that @value{GDBN} can still operate correctly, this minimum is
12669 currently 16 bytes.
12671 The limit applies to the results of some subexpressions as well as to
12672 complete expressions. For example, an expression denoting a simple
12673 integer component, such as @code{x.y.z}, may fail if the size of
12674 @var{x.y} is dynamic and exceeds @var{bytes}. On the other hand,
12675 @value{GDBN} is sometimes clever; the expression @code{A[i]}, where
12676 @var{A} is an array variable with non-constant size, will generally
12677 succeed regardless of the bounds on @var{A}, as long as the component
12678 size is less than @var{bytes}.
12680 The default value of @code{max-value-size} is currently 64k.
12682 @kindex show max-value-size
12683 @item show max-value-size
12684 Show the maximum size of memory, in bytes, that @value{GDBN} will
12685 allocate for the contents of a value.
12688 @node Optimized Code
12689 @chapter Debugging Optimized Code
12690 @cindex optimized code, debugging
12691 @cindex debugging optimized code
12693 Almost all compilers support optimization. With optimization
12694 disabled, the compiler generates assembly code that corresponds
12695 directly to your source code, in a simplistic way. As the compiler
12696 applies more powerful optimizations, the generated assembly code
12697 diverges from your original source code. With help from debugging
12698 information generated by the compiler, @value{GDBN} can map from
12699 the running program back to constructs from your original source.
12701 @value{GDBN} is more accurate with optimization disabled. If you
12702 can recompile without optimization, it is easier to follow the
12703 progress of your program during debugging. But, there are many cases
12704 where you may need to debug an optimized version.
12706 When you debug a program compiled with @samp{-g -O}, remember that the
12707 optimizer has rearranged your code; the debugger shows you what is
12708 really there. Do not be too surprised when the execution path does not
12709 exactly match your source file! An extreme example: if you define a
12710 variable, but never use it, @value{GDBN} never sees that
12711 variable---because the compiler optimizes it out of existence.
12713 Some things do not work as well with @samp{-g -O} as with just
12714 @samp{-g}, particularly on machines with instruction scheduling. If in
12715 doubt, recompile with @samp{-g} alone, and if this fixes the problem,
12716 please report it to us as a bug (including a test case!).
12717 @xref{Variables}, for more information about debugging optimized code.
12720 * Inline Functions:: How @value{GDBN} presents inlining
12721 * Tail Call Frames:: @value{GDBN} analysis of jumps to functions
12724 @node Inline Functions
12725 @section Inline Functions
12726 @cindex inline functions, debugging
12728 @dfn{Inlining} is an optimization that inserts a copy of the function
12729 body directly at each call site, instead of jumping to a shared
12730 routine. @value{GDBN} displays inlined functions just like
12731 non-inlined functions. They appear in backtraces. You can view their
12732 arguments and local variables, step into them with @code{step}, skip
12733 them with @code{next}, and escape from them with @code{finish}.
12734 You can check whether a function was inlined by using the
12735 @code{info frame} command.
12737 For @value{GDBN} to support inlined functions, the compiler must
12738 record information about inlining in the debug information ---
12739 @value{NGCC} using the @sc{dwarf 2} format does this, and several
12740 other compilers do also. @value{GDBN} only supports inlined functions
12741 when using @sc{dwarf 2}. Versions of @value{NGCC} before 4.1
12742 do not emit two required attributes (@samp{DW_AT_call_file} and
12743 @samp{DW_AT_call_line}); @value{GDBN} does not display inlined
12744 function calls with earlier versions of @value{NGCC}. It instead
12745 displays the arguments and local variables of inlined functions as
12746 local variables in the caller.
12748 The body of an inlined function is directly included at its call site;
12749 unlike a non-inlined function, there are no instructions devoted to
12750 the call. @value{GDBN} still pretends that the call site and the
12751 start of the inlined function are different instructions. Stepping to
12752 the call site shows the call site, and then stepping again shows
12753 the first line of the inlined function, even though no additional
12754 instructions are executed.
12756 This makes source-level debugging much clearer; you can see both the
12757 context of the call and then the effect of the call. Only stepping by
12758 a single instruction using @code{stepi} or @code{nexti} does not do
12759 this; single instruction steps always show the inlined body.
12761 There are some ways that @value{GDBN} does not pretend that inlined
12762 function calls are the same as normal calls:
12766 Setting breakpoints at the call site of an inlined function may not
12767 work, because the call site does not contain any code. @value{GDBN}
12768 may incorrectly move the breakpoint to the next line of the enclosing
12769 function, after the call. This limitation will be removed in a future
12770 version of @value{GDBN}; until then, set a breakpoint on an earlier line
12771 or inside the inlined function instead.
12774 @value{GDBN} cannot locate the return value of inlined calls after
12775 using the @code{finish} command. This is a limitation of compiler-generated
12776 debugging information; after @code{finish}, you can step to the next line
12777 and print a variable where your program stored the return value.
12781 @node Tail Call Frames
12782 @section Tail Call Frames
12783 @cindex tail call frames, debugging
12785 Function @code{B} can call function @code{C} in its very last statement. In
12786 unoptimized compilation the call of @code{C} is immediately followed by return
12787 instruction at the end of @code{B} code. Optimizing compiler may replace the
12788 call and return in function @code{B} into one jump to function @code{C}
12789 instead. Such use of a jump instruction is called @dfn{tail call}.
12791 During execution of function @code{C}, there will be no indication in the
12792 function call stack frames that it was tail-called from @code{B}. If function
12793 @code{A} regularly calls function @code{B} which tail-calls function @code{C},
12794 then @value{GDBN} will see @code{A} as the caller of @code{C}. However, in
12795 some cases @value{GDBN} can determine that @code{C} was tail-called from
12796 @code{B}, and it will then create fictitious call frame for that, with the
12797 return address set up as if @code{B} called @code{C} normally.
12799 This functionality is currently supported only by DWARF 2 debugging format and
12800 the compiler has to produce @samp{DW_TAG_call_site} tags. With
12801 @value{NGCC}, you need to specify @option{-O -g} during compilation, to get
12804 @kbd{info frame} command (@pxref{Frame Info}) will indicate the tail call frame
12805 kind by text @code{tail call frame} such as in this sample @value{GDBN} output:
12809 0x40066b <b(int, double)+11>: jmp 0x400640 <c(int, double)>
12811 Stack level 1, frame at 0x7fffffffda30:
12812 rip = 0x40066d in b (amd64-entry-value.cc:59); saved rip 0x4004c5
12813 tail call frame, caller of frame at 0x7fffffffda30
12814 source language c++.
12815 Arglist at unknown address.
12816 Locals at unknown address, Previous frame's sp is 0x7fffffffda30
12819 The detection of all the possible code path executions can find them ambiguous.
12820 There is no execution history stored (possible @ref{Reverse Execution} is never
12821 used for this purpose) and the last known caller could have reached the known
12822 callee by multiple different jump sequences. In such case @value{GDBN} still
12823 tries to show at least all the unambiguous top tail callers and all the
12824 unambiguous bottom tail calees, if any.
12827 @anchor{set debug entry-values}
12828 @item set debug entry-values
12829 @kindex set debug entry-values
12830 When set to on, enables printing of analysis messages for both frame argument
12831 values at function entry and tail calls. It will show all the possible valid
12832 tail calls code paths it has considered. It will also print the intersection
12833 of them with the final unambiguous (possibly partial or even empty) code path
12836 @item show debug entry-values
12837 @kindex show debug entry-values
12838 Show the current state of analysis messages printing for both frame argument
12839 values at function entry and tail calls.
12842 The analysis messages for tail calls can for example show why the virtual tail
12843 call frame for function @code{c} has not been recognized (due to the indirect
12844 reference by variable @code{x}):
12847 static void __attribute__((noinline, noclone)) c (void);
12848 void (*x) (void) = c;
12849 static void __attribute__((noinline, noclone)) a (void) @{ x++; @}
12850 static void __attribute__((noinline, noclone)) c (void) @{ a (); @}
12851 int main (void) @{ x (); return 0; @}
12853 Breakpoint 1, DW_OP_entry_value resolving cannot find
12854 DW_TAG_call_site 0x40039a in main
12856 3 static void __attribute__((noinline, noclone)) a (void) @{ x++; @}
12859 #1 0x000000000040039a in main () at t.c:5
12862 Another possibility is an ambiguous virtual tail call frames resolution:
12866 static void __attribute__((noinline, noclone)) f (void) @{ i++; @}
12867 static void __attribute__((noinline, noclone)) e (void) @{ f (); @}
12868 static void __attribute__((noinline, noclone)) d (void) @{ f (); @}
12869 static void __attribute__((noinline, noclone)) c (void) @{ d (); @}
12870 static void __attribute__((noinline, noclone)) b (void)
12871 @{ if (i) c (); else e (); @}
12872 static void __attribute__((noinline, noclone)) a (void) @{ b (); @}
12873 int main (void) @{ a (); return 0; @}
12875 tailcall: initial: 0x4004d2(a) 0x4004ce(b) 0x4004b2(c) 0x4004a2(d)
12876 tailcall: compare: 0x4004d2(a) 0x4004cc(b) 0x400492(e)
12877 tailcall: reduced: 0x4004d2(a) |
12880 #1 0x00000000004004d2 in a () at t.c:8
12881 #2 0x0000000000400395 in main () at t.c:9
12884 @set CALLSEQ1A @code{main@value{ARROW}a@value{ARROW}b@value{ARROW}c@value{ARROW}d@value{ARROW}f}
12885 @set CALLSEQ2A @code{main@value{ARROW}a@value{ARROW}b@value{ARROW}e@value{ARROW}f}
12887 @c Convert CALLSEQ#A to CALLSEQ#B depending on HAVE_MAKEINFO_CLICK.
12888 @ifset HAVE_MAKEINFO_CLICK
12889 @set ARROW @click{}
12890 @set CALLSEQ1B @clicksequence{@value{CALLSEQ1A}}
12891 @set CALLSEQ2B @clicksequence{@value{CALLSEQ2A}}
12893 @ifclear HAVE_MAKEINFO_CLICK
12895 @set CALLSEQ1B @value{CALLSEQ1A}
12896 @set CALLSEQ2B @value{CALLSEQ2A}
12899 Frames #0 and #2 are real, #1 is a virtual tail call frame.
12900 The code can have possible execution paths @value{CALLSEQ1B} or
12901 @value{CALLSEQ2B}, @value{GDBN} cannot find which one from the inferior state.
12903 @code{initial:} state shows some random possible calling sequence @value{GDBN}
12904 has found. It then finds another possible calling sequcen - that one is
12905 prefixed by @code{compare:}. The non-ambiguous intersection of these two is
12906 printed as the @code{reduced:} calling sequence. That one could have many
12907 futher @code{compare:} and @code{reduced:} statements as long as there remain
12908 any non-ambiguous sequence entries.
12910 For the frame of function @code{b} in both cases there are different possible
12911 @code{$pc} values (@code{0x4004cc} or @code{0x4004ce}), therefore this frame is
12912 also ambigous. The only non-ambiguous frame is the one for function @code{a},
12913 therefore this one is displayed to the user while the ambiguous frames are
12916 There can be also reasons why printing of frame argument values at function
12921 static void __attribute__((noinline, noclone)) c (int i) @{ v++; @}
12922 static void __attribute__((noinline, noclone)) a (int i);
12923 static void __attribute__((noinline, noclone)) b (int i) @{ a (i); @}
12924 static void __attribute__((noinline, noclone)) a (int i)
12925 @{ if (i) b (i - 1); else c (0); @}
12926 int main (void) @{ a (5); return 0; @}
12929 #0 c (i=i@@entry=0) at t.c:2
12930 #1 0x0000000000400428 in a (DW_OP_entry_value resolving has found
12931 function "a" at 0x400420 can call itself via tail calls
12932 i=<optimized out>) at t.c:6
12933 #2 0x000000000040036e in main () at t.c:7
12936 @value{GDBN} cannot find out from the inferior state if and how many times did
12937 function @code{a} call itself (via function @code{b}) as these calls would be
12938 tail calls. Such tail calls would modify thue @code{i} variable, therefore
12939 @value{GDBN} cannot be sure the value it knows would be right - @value{GDBN}
12940 prints @code{<optimized out>} instead.
12943 @chapter C Preprocessor Macros
12945 Some languages, such as C and C@t{++}, provide a way to define and invoke
12946 ``preprocessor macros'' which expand into strings of tokens.
12947 @value{GDBN} can evaluate expressions containing macro invocations, show
12948 the result of macro expansion, and show a macro's definition, including
12949 where it was defined.
12951 You may need to compile your program specially to provide @value{GDBN}
12952 with information about preprocessor macros. Most compilers do not
12953 include macros in their debugging information, even when you compile
12954 with the @option{-g} flag. @xref{Compilation}.
12956 A program may define a macro at one point, remove that definition later,
12957 and then provide a different definition after that. Thus, at different
12958 points in the program, a macro may have different definitions, or have
12959 no definition at all. If there is a current stack frame, @value{GDBN}
12960 uses the macros in scope at that frame's source code line. Otherwise,
12961 @value{GDBN} uses the macros in scope at the current listing location;
12964 Whenever @value{GDBN} evaluates an expression, it always expands any
12965 macro invocations present in the expression. @value{GDBN} also provides
12966 the following commands for working with macros explicitly.
12970 @kindex macro expand
12971 @cindex macro expansion, showing the results of preprocessor
12972 @cindex preprocessor macro expansion, showing the results of
12973 @cindex expanding preprocessor macros
12974 @item macro expand @var{expression}
12975 @itemx macro exp @var{expression}
12976 Show the results of expanding all preprocessor macro invocations in
12977 @var{expression}. Since @value{GDBN} simply expands macros, but does
12978 not parse the result, @var{expression} need not be a valid expression;
12979 it can be any string of tokens.
12982 @item macro expand-once @var{expression}
12983 @itemx macro exp1 @var{expression}
12984 @cindex expand macro once
12985 @i{(This command is not yet implemented.)} Show the results of
12986 expanding those preprocessor macro invocations that appear explicitly in
12987 @var{expression}. Macro invocations appearing in that expansion are
12988 left unchanged. This command allows you to see the effect of a
12989 particular macro more clearly, without being confused by further
12990 expansions. Since @value{GDBN} simply expands macros, but does not
12991 parse the result, @var{expression} need not be a valid expression; it
12992 can be any string of tokens.
12995 @cindex macro definition, showing
12996 @cindex definition of a macro, showing
12997 @cindex macros, from debug info
12998 @item info macro [-a|-all] [--] @var{macro}
12999 Show the current definition or all definitions of the named @var{macro},
13000 and describe the source location or compiler command-line where that
13001 definition was established. The optional double dash is to signify the end of
13002 argument processing and the beginning of @var{macro} for non C-like macros where
13003 the macro may begin with a hyphen.
13005 @kindex info macros
13006 @item info macros @var{location}
13007 Show all macro definitions that are in effect at the location specified
13008 by @var{location}, and describe the source location or compiler
13009 command-line where those definitions were established.
13011 @kindex macro define
13012 @cindex user-defined macros
13013 @cindex defining macros interactively
13014 @cindex macros, user-defined
13015 @item macro define @var{macro} @var{replacement-list}
13016 @itemx macro define @var{macro}(@var{arglist}) @var{replacement-list}
13017 Introduce a definition for a preprocessor macro named @var{macro},
13018 invocations of which are replaced by the tokens given in
13019 @var{replacement-list}. The first form of this command defines an
13020 ``object-like'' macro, which takes no arguments; the second form
13021 defines a ``function-like'' macro, which takes the arguments given in
13024 A definition introduced by this command is in scope in every
13025 expression evaluated in @value{GDBN}, until it is removed with the
13026 @code{macro undef} command, described below. The definition overrides
13027 all definitions for @var{macro} present in the program being debugged,
13028 as well as any previous user-supplied definition.
13030 @kindex macro undef
13031 @item macro undef @var{macro}
13032 Remove any user-supplied definition for the macro named @var{macro}.
13033 This command only affects definitions provided with the @code{macro
13034 define} command, described above; it cannot remove definitions present
13035 in the program being debugged.
13039 List all the macros defined using the @code{macro define} command.
13042 @cindex macros, example of debugging with
13043 Here is a transcript showing the above commands in action. First, we
13044 show our source files:
13049 #include "sample.h"
13052 #define ADD(x) (M + x)
13057 printf ("Hello, world!\n");
13059 printf ("We're so creative.\n");
13061 printf ("Goodbye, world!\n");
13068 Now, we compile the program using the @sc{gnu} C compiler,
13069 @value{NGCC}. We pass the @option{-gdwarf-2}@footnote{This is the
13070 minimum. Recent versions of @value{NGCC} support @option{-gdwarf-3}
13071 and @option{-gdwarf-4}; we recommend always choosing the most recent
13072 version of DWARF.} @emph{and} @option{-g3} flags to ensure the compiler
13073 includes information about preprocessor macros in the debugging
13077 $ gcc -gdwarf-2 -g3 sample.c -o sample
13081 Now, we start @value{GDBN} on our sample program:
13085 GNU gdb 2002-05-06-cvs
13086 Copyright 2002 Free Software Foundation, Inc.
13087 GDB is free software, @dots{}
13091 We can expand macros and examine their definitions, even when the
13092 program is not running. @value{GDBN} uses the current listing position
13093 to decide which macro definitions are in scope:
13096 (@value{GDBP}) list main
13099 5 #define ADD(x) (M + x)
13104 10 printf ("Hello, world!\n");
13106 12 printf ("We're so creative.\n");
13107 (@value{GDBP}) info macro ADD
13108 Defined at /home/jimb/gdb/macros/play/sample.c:5
13109 #define ADD(x) (M + x)
13110 (@value{GDBP}) info macro Q
13111 Defined at /home/jimb/gdb/macros/play/sample.h:1
13112 included at /home/jimb/gdb/macros/play/sample.c:2
13114 (@value{GDBP}) macro expand ADD(1)
13115 expands to: (42 + 1)
13116 (@value{GDBP}) macro expand-once ADD(1)
13117 expands to: once (M + 1)
13121 In the example above, note that @code{macro expand-once} expands only
13122 the macro invocation explicit in the original text --- the invocation of
13123 @code{ADD} --- but does not expand the invocation of the macro @code{M},
13124 which was introduced by @code{ADD}.
13126 Once the program is running, @value{GDBN} uses the macro definitions in
13127 force at the source line of the current stack frame:
13130 (@value{GDBP}) break main
13131 Breakpoint 1 at 0x8048370: file sample.c, line 10.
13133 Starting program: /home/jimb/gdb/macros/play/sample
13135 Breakpoint 1, main () at sample.c:10
13136 10 printf ("Hello, world!\n");
13140 At line 10, the definition of the macro @code{N} at line 9 is in force:
13143 (@value{GDBP}) info macro N
13144 Defined at /home/jimb/gdb/macros/play/sample.c:9
13146 (@value{GDBP}) macro expand N Q M
13147 expands to: 28 < 42
13148 (@value{GDBP}) print N Q M
13153 As we step over directives that remove @code{N}'s definition, and then
13154 give it a new definition, @value{GDBN} finds the definition (or lack
13155 thereof) in force at each point:
13158 (@value{GDBP}) next
13160 12 printf ("We're so creative.\n");
13161 (@value{GDBP}) info macro N
13162 The symbol `N' has no definition as a C/C++ preprocessor macro
13163 at /home/jimb/gdb/macros/play/sample.c:12
13164 (@value{GDBP}) next
13166 14 printf ("Goodbye, world!\n");
13167 (@value{GDBP}) info macro N
13168 Defined at /home/jimb/gdb/macros/play/sample.c:13
13170 (@value{GDBP}) macro expand N Q M
13171 expands to: 1729 < 42
13172 (@value{GDBP}) print N Q M
13177 In addition to source files, macros can be defined on the compilation command
13178 line using the @option{-D@var{name}=@var{value}} syntax. For macros defined in
13179 such a way, @value{GDBN} displays the location of their definition as line zero
13180 of the source file submitted to the compiler.
13183 (@value{GDBP}) info macro __STDC__
13184 Defined at /home/jimb/gdb/macros/play/sample.c:0
13191 @chapter Tracepoints
13192 @c This chapter is based on the documentation written by Michael
13193 @c Snyder, David Taylor, Jim Blandy, and Elena Zannoni.
13195 @cindex tracepoints
13196 In some applications, it is not feasible for the debugger to interrupt
13197 the program's execution long enough for the developer to learn
13198 anything helpful about its behavior. If the program's correctness
13199 depends on its real-time behavior, delays introduced by a debugger
13200 might cause the program to change its behavior drastically, or perhaps
13201 fail, even when the code itself is correct. It is useful to be able
13202 to observe the program's behavior without interrupting it.
13204 Using @value{GDBN}'s @code{trace} and @code{collect} commands, you can
13205 specify locations in the program, called @dfn{tracepoints}, and
13206 arbitrary expressions to evaluate when those tracepoints are reached.
13207 Later, using the @code{tfind} command, you can examine the values
13208 those expressions had when the program hit the tracepoints. The
13209 expressions may also denote objects in memory---structures or arrays,
13210 for example---whose values @value{GDBN} should record; while visiting
13211 a particular tracepoint, you may inspect those objects as if they were
13212 in memory at that moment. However, because @value{GDBN} records these
13213 values without interacting with you, it can do so quickly and
13214 unobtrusively, hopefully not disturbing the program's behavior.
13216 The tracepoint facility is currently available only for remote
13217 targets. @xref{Targets}. In addition, your remote target must know
13218 how to collect trace data. This functionality is implemented in the
13219 remote stub; however, none of the stubs distributed with @value{GDBN}
13220 support tracepoints as of this writing. The format of the remote
13221 packets used to implement tracepoints are described in @ref{Tracepoint
13224 It is also possible to get trace data from a file, in a manner reminiscent
13225 of corefiles; you specify the filename, and use @code{tfind} to search
13226 through the file. @xref{Trace Files}, for more details.
13228 This chapter describes the tracepoint commands and features.
13231 * Set Tracepoints::
13232 * Analyze Collected Data::
13233 * Tracepoint Variables::
13237 @node Set Tracepoints
13238 @section Commands to Set Tracepoints
13240 Before running such a @dfn{trace experiment}, an arbitrary number of
13241 tracepoints can be set. A tracepoint is actually a special type of
13242 breakpoint (@pxref{Set Breaks}), so you can manipulate it using
13243 standard breakpoint commands. For instance, as with breakpoints,
13244 tracepoint numbers are successive integers starting from one, and many
13245 of the commands associated with tracepoints take the tracepoint number
13246 as their argument, to identify which tracepoint to work on.
13248 For each tracepoint, you can specify, in advance, some arbitrary set
13249 of data that you want the target to collect in the trace buffer when
13250 it hits that tracepoint. The collected data can include registers,
13251 local variables, or global data. Later, you can use @value{GDBN}
13252 commands to examine the values these data had at the time the
13253 tracepoint was hit.
13255 Tracepoints do not support every breakpoint feature. Ignore counts on
13256 tracepoints have no effect, and tracepoints cannot run @value{GDBN}
13257 commands when they are hit. Tracepoints may not be thread-specific
13260 @cindex fast tracepoints
13261 Some targets may support @dfn{fast tracepoints}, which are inserted in
13262 a different way (such as with a jump instead of a trap), that is
13263 faster but possibly restricted in where they may be installed.
13265 @cindex static tracepoints
13266 @cindex markers, static tracepoints
13267 @cindex probing markers, static tracepoints
13268 Regular and fast tracepoints are dynamic tracing facilities, meaning
13269 that they can be used to insert tracepoints at (almost) any location
13270 in the target. Some targets may also support controlling @dfn{static
13271 tracepoints} from @value{GDBN}. With static tracing, a set of
13272 instrumentation points, also known as @dfn{markers}, are embedded in
13273 the target program, and can be activated or deactivated by name or
13274 address. These are usually placed at locations which facilitate
13275 investigating what the target is actually doing. @value{GDBN}'s
13276 support for static tracing includes being able to list instrumentation
13277 points, and attach them with @value{GDBN} defined high level
13278 tracepoints that expose the whole range of convenience of
13279 @value{GDBN}'s tracepoints support. Namely, support for collecting
13280 registers values and values of global or local (to the instrumentation
13281 point) variables; tracepoint conditions and trace state variables.
13282 The act of installing a @value{GDBN} static tracepoint on an
13283 instrumentation point, or marker, is referred to as @dfn{probing} a
13284 static tracepoint marker.
13286 @code{gdbserver} supports tracepoints on some target systems.
13287 @xref{Server,,Tracepoints support in @code{gdbserver}}.
13289 This section describes commands to set tracepoints and associated
13290 conditions and actions.
13293 * Create and Delete Tracepoints::
13294 * Enable and Disable Tracepoints::
13295 * Tracepoint Passcounts::
13296 * Tracepoint Conditions::
13297 * Trace State Variables::
13298 * Tracepoint Actions::
13299 * Listing Tracepoints::
13300 * Listing Static Tracepoint Markers::
13301 * Starting and Stopping Trace Experiments::
13302 * Tracepoint Restrictions::
13305 @node Create and Delete Tracepoints
13306 @subsection Create and Delete Tracepoints
13309 @cindex set tracepoint
13311 @item trace @var{location}
13312 The @code{trace} command is very similar to the @code{break} command.
13313 Its argument @var{location} can be any valid location.
13314 @xref{Specify Location}. The @code{trace} command defines a tracepoint,
13315 which is a point in the target program where the debugger will briefly stop,
13316 collect some data, and then allow the program to continue. Setting a tracepoint
13317 or changing its actions takes effect immediately if the remote stub
13318 supports the @samp{InstallInTrace} feature (@pxref{install tracepoint
13320 If remote stub doesn't support the @samp{InstallInTrace} feature, all
13321 these changes don't take effect until the next @code{tstart}
13322 command, and once a trace experiment is running, further changes will
13323 not have any effect until the next trace experiment starts. In addition,
13324 @value{GDBN} supports @dfn{pending tracepoints}---tracepoints whose
13325 address is not yet resolved. (This is similar to pending breakpoints.)
13326 Pending tracepoints are not downloaded to the target and not installed
13327 until they are resolved. The resolution of pending tracepoints requires
13328 @value{GDBN} support---when debugging with the remote target, and
13329 @value{GDBN} disconnects from the remote stub (@pxref{disconnected
13330 tracing}), pending tracepoints can not be resolved (and downloaded to
13331 the remote stub) while @value{GDBN} is disconnected.
13333 Here are some examples of using the @code{trace} command:
13336 (@value{GDBP}) @b{trace foo.c:121} // a source file and line number
13338 (@value{GDBP}) @b{trace +2} // 2 lines forward
13340 (@value{GDBP}) @b{trace my_function} // first source line of function
13342 (@value{GDBP}) @b{trace *my_function} // EXACT start address of function
13344 (@value{GDBP}) @b{trace *0x2117c4} // an address
13348 You can abbreviate @code{trace} as @code{tr}.
13350 @item trace @var{location} if @var{cond}
13351 Set a tracepoint with condition @var{cond}; evaluate the expression
13352 @var{cond} each time the tracepoint is reached, and collect data only
13353 if the value is nonzero---that is, if @var{cond} evaluates as true.
13354 @xref{Tracepoint Conditions, ,Tracepoint Conditions}, for more
13355 information on tracepoint conditions.
13357 @item ftrace @var{location} [ if @var{cond} ]
13358 @cindex set fast tracepoint
13359 @cindex fast tracepoints, setting
13361 The @code{ftrace} command sets a fast tracepoint. For targets that
13362 support them, fast tracepoints will use a more efficient but possibly
13363 less general technique to trigger data collection, such as a jump
13364 instruction instead of a trap, or some sort of hardware support. It
13365 may not be possible to create a fast tracepoint at the desired
13366 location, in which case the command will exit with an explanatory
13369 @value{GDBN} handles arguments to @code{ftrace} exactly as for
13372 On 32-bit x86-architecture systems, fast tracepoints normally need to
13373 be placed at an instruction that is 5 bytes or longer, but can be
13374 placed at 4-byte instructions if the low 64K of memory of the target
13375 program is available to install trampolines. Some Unix-type systems,
13376 such as @sc{gnu}/Linux, exclude low addresses from the program's
13377 address space; but for instance with the Linux kernel it is possible
13378 to let @value{GDBN} use this area by doing a @command{sysctl} command
13379 to set the @code{mmap_min_addr} kernel parameter, as in
13382 sudo sysctl -w vm.mmap_min_addr=32768
13386 which sets the low address to 32K, which leaves plenty of room for
13387 trampolines. The minimum address should be set to a page boundary.
13389 @item strace @var{location} [ if @var{cond} ]
13390 @cindex set static tracepoint
13391 @cindex static tracepoints, setting
13392 @cindex probe static tracepoint marker
13394 The @code{strace} command sets a static tracepoint. For targets that
13395 support it, setting a static tracepoint probes a static
13396 instrumentation point, or marker, found at @var{location}. It may not
13397 be possible to set a static tracepoint at the desired location, in
13398 which case the command will exit with an explanatory message.
13400 @value{GDBN} handles arguments to @code{strace} exactly as for
13401 @code{trace}, with the addition that the user can also specify
13402 @code{-m @var{marker}} as @var{location}. This probes the marker
13403 identified by the @var{marker} string identifier. This identifier
13404 depends on the static tracepoint backend library your program is
13405 using. You can find all the marker identifiers in the @samp{ID} field
13406 of the @code{info static-tracepoint-markers} command output.
13407 @xref{Listing Static Tracepoint Markers,,Listing Static Tracepoint
13408 Markers}. For example, in the following small program using the UST
13414 trace_mark(ust, bar33, "str %s", "FOOBAZ");
13419 the marker id is composed of joining the first two arguments to the
13420 @code{trace_mark} call with a slash, which translates to:
13423 (@value{GDBP}) info static-tracepoint-markers
13424 Cnt Enb ID Address What
13425 1 n ust/bar33 0x0000000000400ddc in main at stexample.c:22
13431 so you may probe the marker above with:
13434 (@value{GDBP}) strace -m ust/bar33
13437 Static tracepoints accept an extra collect action --- @code{collect
13438 $_sdata}. This collects arbitrary user data passed in the probe point
13439 call to the tracing library. In the UST example above, you'll see
13440 that the third argument to @code{trace_mark} is a printf-like format
13441 string. The user data is then the result of running that formating
13442 string against the following arguments. Note that @code{info
13443 static-tracepoint-markers} command output lists that format string in
13444 the @samp{Data:} field.
13446 You can inspect this data when analyzing the trace buffer, by printing
13447 the $_sdata variable like any other variable available to
13448 @value{GDBN}. @xref{Tracepoint Actions,,Tracepoint Action Lists}.
13451 @cindex last tracepoint number
13452 @cindex recent tracepoint number
13453 @cindex tracepoint number
13454 The convenience variable @code{$tpnum} records the tracepoint number
13455 of the most recently set tracepoint.
13457 @kindex delete tracepoint
13458 @cindex tracepoint deletion
13459 @item delete tracepoint @r{[}@var{num}@r{]}
13460 Permanently delete one or more tracepoints. With no argument, the
13461 default is to delete all tracepoints. Note that the regular
13462 @code{delete} command can remove tracepoints also.
13467 (@value{GDBP}) @b{delete trace 1 2 3} // remove three tracepoints
13469 (@value{GDBP}) @b{delete trace} // remove all tracepoints
13473 You can abbreviate this command as @code{del tr}.
13476 @node Enable and Disable Tracepoints
13477 @subsection Enable and Disable Tracepoints
13479 These commands are deprecated; they are equivalent to plain @code{disable} and @code{enable}.
13482 @kindex disable tracepoint
13483 @item disable tracepoint @r{[}@var{num}@r{]}
13484 Disable tracepoint @var{num}, or all tracepoints if no argument
13485 @var{num} is given. A disabled tracepoint will have no effect during
13486 a trace experiment, but it is not forgotten. You can re-enable
13487 a disabled tracepoint using the @code{enable tracepoint} command.
13488 If the command is issued during a trace experiment and the debug target
13489 has support for disabling tracepoints during a trace experiment, then the
13490 change will be effective immediately. Otherwise, it will be applied to the
13491 next trace experiment.
13493 @kindex enable tracepoint
13494 @item enable tracepoint @r{[}@var{num}@r{]}
13495 Enable tracepoint @var{num}, or all tracepoints. If this command is
13496 issued during a trace experiment and the debug target supports enabling
13497 tracepoints during a trace experiment, then the enabled tracepoints will
13498 become effective immediately. Otherwise, they will become effective the
13499 next time a trace experiment is run.
13502 @node Tracepoint Passcounts
13503 @subsection Tracepoint Passcounts
13507 @cindex tracepoint pass count
13508 @item passcount @r{[}@var{n} @r{[}@var{num}@r{]]}
13509 Set the @dfn{passcount} of a tracepoint. The passcount is a way to
13510 automatically stop a trace experiment. If a tracepoint's passcount is
13511 @var{n}, then the trace experiment will be automatically stopped on
13512 the @var{n}'th time that tracepoint is hit. If the tracepoint number
13513 @var{num} is not specified, the @code{passcount} command sets the
13514 passcount of the most recently defined tracepoint. If no passcount is
13515 given, the trace experiment will run until stopped explicitly by the
13521 (@value{GDBP}) @b{passcount 5 2} // Stop on the 5th execution of
13522 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// tracepoint 2}
13524 (@value{GDBP}) @b{passcount 12} // Stop on the 12th execution of the
13525 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// most recently defined tracepoint.}
13526 (@value{GDBP}) @b{trace foo}
13527 (@value{GDBP}) @b{pass 3}
13528 (@value{GDBP}) @b{trace bar}
13529 (@value{GDBP}) @b{pass 2}
13530 (@value{GDBP}) @b{trace baz}
13531 (@value{GDBP}) @b{pass 1} // Stop tracing when foo has been
13532 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// executed 3 times OR when bar has}
13533 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// been executed 2 times}
13534 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// OR when baz has been executed 1 time.}
13538 @node Tracepoint Conditions
13539 @subsection Tracepoint Conditions
13540 @cindex conditional tracepoints
13541 @cindex tracepoint conditions
13543 The simplest sort of tracepoint collects data every time your program
13544 reaches a specified place. You can also specify a @dfn{condition} for
13545 a tracepoint. A condition is just a Boolean expression in your
13546 programming language (@pxref{Expressions, ,Expressions}). A
13547 tracepoint with a condition evaluates the expression each time your
13548 program reaches it, and data collection happens only if the condition
13551 Tracepoint conditions can be specified when a tracepoint is set, by
13552 using @samp{if} in the arguments to the @code{trace} command.
13553 @xref{Create and Delete Tracepoints, ,Setting Tracepoints}. They can
13554 also be set or changed at any time with the @code{condition} command,
13555 just as with breakpoints.
13557 Unlike breakpoint conditions, @value{GDBN} does not actually evaluate
13558 the conditional expression itself. Instead, @value{GDBN} encodes the
13559 expression into an agent expression (@pxref{Agent Expressions})
13560 suitable for execution on the target, independently of @value{GDBN}.
13561 Global variables become raw memory locations, locals become stack
13562 accesses, and so forth.
13564 For instance, suppose you have a function that is usually called
13565 frequently, but should not be called after an error has occurred. You
13566 could use the following tracepoint command to collect data about calls
13567 of that function that happen while the error code is propagating
13568 through the program; an unconditional tracepoint could end up
13569 collecting thousands of useless trace frames that you would have to
13573 (@value{GDBP}) @kbd{trace normal_operation if errcode > 0}
13576 @node Trace State Variables
13577 @subsection Trace State Variables
13578 @cindex trace state variables
13580 A @dfn{trace state variable} is a special type of variable that is
13581 created and managed by target-side code. The syntax is the same as
13582 that for GDB's convenience variables (a string prefixed with ``$''),
13583 but they are stored on the target. They must be created explicitly,
13584 using a @code{tvariable} command. They are always 64-bit signed
13587 Trace state variables are remembered by @value{GDBN}, and downloaded
13588 to the target along with tracepoint information when the trace
13589 experiment starts. There are no intrinsic limits on the number of
13590 trace state variables, beyond memory limitations of the target.
13592 @cindex convenience variables, and trace state variables
13593 Although trace state variables are managed by the target, you can use
13594 them in print commands and expressions as if they were convenience
13595 variables; @value{GDBN} will get the current value from the target
13596 while the trace experiment is running. Trace state variables share
13597 the same namespace as other ``$'' variables, which means that you
13598 cannot have trace state variables with names like @code{$23} or
13599 @code{$pc}, nor can you have a trace state variable and a convenience
13600 variable with the same name.
13604 @item tvariable $@var{name} [ = @var{expression} ]
13606 The @code{tvariable} command creates a new trace state variable named
13607 @code{$@var{name}}, and optionally gives it an initial value of
13608 @var{expression}. The @var{expression} is evaluated when this command is
13609 entered; the result will be converted to an integer if possible,
13610 otherwise @value{GDBN} will report an error. A subsequent
13611 @code{tvariable} command specifying the same name does not create a
13612 variable, but instead assigns the supplied initial value to the
13613 existing variable of that name, overwriting any previous initial
13614 value. The default initial value is 0.
13616 @item info tvariables
13617 @kindex info tvariables
13618 List all the trace state variables along with their initial values.
13619 Their current values may also be displayed, if the trace experiment is
13622 @item delete tvariable @r{[} $@var{name} @dots{} @r{]}
13623 @kindex delete tvariable
13624 Delete the given trace state variables, or all of them if no arguments
13629 @node Tracepoint Actions
13630 @subsection Tracepoint Action Lists
13634 @cindex tracepoint actions
13635 @item actions @r{[}@var{num}@r{]}
13636 This command will prompt for a list of actions to be taken when the
13637 tracepoint is hit. If the tracepoint number @var{num} is not
13638 specified, this command sets the actions for the one that was most
13639 recently defined (so that you can define a tracepoint and then say
13640 @code{actions} without bothering about its number). You specify the
13641 actions themselves on the following lines, one action at a time, and
13642 terminate the actions list with a line containing just @code{end}. So
13643 far, the only defined actions are @code{collect}, @code{teval}, and
13644 @code{while-stepping}.
13646 @code{actions} is actually equivalent to @code{commands} (@pxref{Break
13647 Commands, ,Breakpoint Command Lists}), except that only the defined
13648 actions are allowed; any other @value{GDBN} command is rejected.
13650 @cindex remove actions from a tracepoint
13651 To remove all actions from a tracepoint, type @samp{actions @var{num}}
13652 and follow it immediately with @samp{end}.
13655 (@value{GDBP}) @b{collect @var{data}} // collect some data
13657 (@value{GDBP}) @b{while-stepping 5} // single-step 5 times, collect data
13659 (@value{GDBP}) @b{end} // signals the end of actions.
13662 In the following example, the action list begins with @code{collect}
13663 commands indicating the things to be collected when the tracepoint is
13664 hit. Then, in order to single-step and collect additional data
13665 following the tracepoint, a @code{while-stepping} command is used,
13666 followed by the list of things to be collected after each step in a
13667 sequence of single steps. The @code{while-stepping} command is
13668 terminated by its own separate @code{end} command. Lastly, the action
13669 list is terminated by an @code{end} command.
13672 (@value{GDBP}) @b{trace foo}
13673 (@value{GDBP}) @b{actions}
13674 Enter actions for tracepoint 1, one per line:
13677 > while-stepping 12
13678 > collect $pc, arr[i]
13683 @kindex collect @r{(tracepoints)}
13684 @item collect@r{[}/@var{mods}@r{]} @var{expr1}, @var{expr2}, @dots{}
13685 Collect values of the given expressions when the tracepoint is hit.
13686 This command accepts a comma-separated list of any valid expressions.
13687 In addition to global, static, or local variables, the following
13688 special arguments are supported:
13692 Collect all registers.
13695 Collect all function arguments.
13698 Collect all local variables.
13701 Collect the return address. This is helpful if you want to see more
13704 @emph{Note:} The return address location can not always be reliably
13705 determined up front, and the wrong address / registers may end up
13706 collected instead. On some architectures the reliability is higher
13707 for tracepoints at function entry, while on others it's the opposite.
13708 When this happens, backtracing will stop because the return address is
13709 found unavailable (unless another collect rule happened to match it).
13712 Collects the number of arguments from the static probe at which the
13713 tracepoint is located.
13714 @xref{Static Probe Points}.
13716 @item $_probe_arg@var{n}
13717 @var{n} is an integer between 0 and 11. Collects the @var{n}th argument
13718 from the static probe at which the tracepoint is located.
13719 @xref{Static Probe Points}.
13722 @vindex $_sdata@r{, collect}
13723 Collect static tracepoint marker specific data. Only available for
13724 static tracepoints. @xref{Tracepoint Actions,,Tracepoint Action
13725 Lists}. On the UST static tracepoints library backend, an
13726 instrumentation point resembles a @code{printf} function call. The
13727 tracing library is able to collect user specified data formatted to a
13728 character string using the format provided by the programmer that
13729 instrumented the program. Other backends have similar mechanisms.
13730 Here's an example of a UST marker call:
13733 const char master_name[] = "$your_name";
13734 trace_mark(channel1, marker1, "hello %s", master_name)
13737 In this case, collecting @code{$_sdata} collects the string
13738 @samp{hello $yourname}. When analyzing the trace buffer, you can
13739 inspect @samp{$_sdata} like any other variable available to
13743 You can give several consecutive @code{collect} commands, each one
13744 with a single argument, or one @code{collect} command with several
13745 arguments separated by commas; the effect is the same.
13747 The optional @var{mods} changes the usual handling of the arguments.
13748 @code{s} requests that pointers to chars be handled as strings, in
13749 particular collecting the contents of the memory being pointed at, up
13750 to the first zero. The upper bound is by default the value of the
13751 @code{print elements} variable; if @code{s} is followed by a decimal
13752 number, that is the upper bound instead. So for instance
13753 @samp{collect/s25 mystr} collects as many as 25 characters at
13756 The command @code{info scope} (@pxref{Symbols, info scope}) is
13757 particularly useful for figuring out what data to collect.
13759 @kindex teval @r{(tracepoints)}
13760 @item teval @var{expr1}, @var{expr2}, @dots{}
13761 Evaluate the given expressions when the tracepoint is hit. This
13762 command accepts a comma-separated list of expressions. The results
13763 are discarded, so this is mainly useful for assigning values to trace
13764 state variables (@pxref{Trace State Variables}) without adding those
13765 values to the trace buffer, as would be the case if the @code{collect}
13768 @kindex while-stepping @r{(tracepoints)}
13769 @item while-stepping @var{n}
13770 Perform @var{n} single-step instruction traces after the tracepoint,
13771 collecting new data after each step. The @code{while-stepping}
13772 command is followed by the list of what to collect while stepping
13773 (followed by its own @code{end} command):
13776 > while-stepping 12
13777 > collect $regs, myglobal
13783 Note that @code{$pc} is not automatically collected by
13784 @code{while-stepping}; you need to explicitly collect that register if
13785 you need it. You may abbreviate @code{while-stepping} as @code{ws} or
13788 @item set default-collect @var{expr1}, @var{expr2}, @dots{}
13789 @kindex set default-collect
13790 @cindex default collection action
13791 This variable is a list of expressions to collect at each tracepoint
13792 hit. It is effectively an additional @code{collect} action prepended
13793 to every tracepoint action list. The expressions are parsed
13794 individually for each tracepoint, so for instance a variable named
13795 @code{xyz} may be interpreted as a global for one tracepoint, and a
13796 local for another, as appropriate to the tracepoint's location.
13798 @item show default-collect
13799 @kindex show default-collect
13800 Show the list of expressions that are collected by default at each
13805 @node Listing Tracepoints
13806 @subsection Listing Tracepoints
13809 @kindex info tracepoints @r{[}@var{n}@dots{}@r{]}
13810 @kindex info tp @r{[}@var{n}@dots{}@r{]}
13811 @cindex information about tracepoints
13812 @item info tracepoints @r{[}@var{num}@dots{}@r{]}
13813 Display information about the tracepoint @var{num}. If you don't
13814 specify a tracepoint number, displays information about all the
13815 tracepoints defined so far. The format is similar to that used for
13816 @code{info breakpoints}; in fact, @code{info tracepoints} is the same
13817 command, simply restricting itself to tracepoints.
13819 A tracepoint's listing may include additional information specific to
13824 its passcount as given by the @code{passcount @var{n}} command
13827 the state about installed on target of each location
13831 (@value{GDBP}) @b{info trace}
13832 Num Type Disp Enb Address What
13833 1 tracepoint keep y 0x0804ab57 in foo() at main.cxx:7
13835 collect globfoo, $regs
13840 2 tracepoint keep y <MULTIPLE>
13842 2.1 y 0x0804859c in func4 at change-loc.h:35
13843 installed on target
13844 2.2 y 0xb7ffc480 in func4 at change-loc.h:35
13845 installed on target
13846 2.3 y <PENDING> set_tracepoint
13847 3 tracepoint keep y 0x080485b1 in foo at change-loc.c:29
13848 not installed on target
13853 This command can be abbreviated @code{info tp}.
13856 @node Listing Static Tracepoint Markers
13857 @subsection Listing Static Tracepoint Markers
13860 @kindex info static-tracepoint-markers
13861 @cindex information about static tracepoint markers
13862 @item info static-tracepoint-markers
13863 Display information about all static tracepoint markers defined in the
13866 For each marker, the following columns are printed:
13870 An incrementing counter, output to help readability. This is not a
13873 The marker ID, as reported by the target.
13874 @item Enabled or Disabled
13875 Probed markers are tagged with @samp{y}. @samp{n} identifies marks
13876 that are not enabled.
13878 Where the marker is in your program, as a memory address.
13880 Where the marker is in the source for your program, as a file and line
13881 number. If the debug information included in the program does not
13882 allow @value{GDBN} to locate the source of the marker, this column
13883 will be left blank.
13887 In addition, the following information may be printed for each marker:
13891 User data passed to the tracing library by the marker call. In the
13892 UST backend, this is the format string passed as argument to the
13894 @item Static tracepoints probing the marker
13895 The list of static tracepoints attached to the marker.
13899 (@value{GDBP}) info static-tracepoint-markers
13900 Cnt ID Enb Address What
13901 1 ust/bar2 y 0x0000000000400e1a in main at stexample.c:25
13902 Data: number1 %d number2 %d
13903 Probed by static tracepoints: #2
13904 2 ust/bar33 n 0x0000000000400c87 in main at stexample.c:24
13910 @node Starting and Stopping Trace Experiments
13911 @subsection Starting and Stopping Trace Experiments
13914 @kindex tstart [ @var{notes} ]
13915 @cindex start a new trace experiment
13916 @cindex collected data discarded
13918 This command starts the trace experiment, and begins collecting data.
13919 It has the side effect of discarding all the data collected in the
13920 trace buffer during the previous trace experiment. If any arguments
13921 are supplied, they are taken as a note and stored with the trace
13922 experiment's state. The notes may be arbitrary text, and are
13923 especially useful with disconnected tracing in a multi-user context;
13924 the notes can explain what the trace is doing, supply user contact
13925 information, and so forth.
13927 @kindex tstop [ @var{notes} ]
13928 @cindex stop a running trace experiment
13930 This command stops the trace experiment. If any arguments are
13931 supplied, they are recorded with the experiment as a note. This is
13932 useful if you are stopping a trace started by someone else, for
13933 instance if the trace is interfering with the system's behavior and
13934 needs to be stopped quickly.
13936 @strong{Note}: a trace experiment and data collection may stop
13937 automatically if any tracepoint's passcount is reached
13938 (@pxref{Tracepoint Passcounts}), or if the trace buffer becomes full.
13941 @cindex status of trace data collection
13942 @cindex trace experiment, status of
13944 This command displays the status of the current trace data
13948 Here is an example of the commands we described so far:
13951 (@value{GDBP}) @b{trace gdb_c_test}
13952 (@value{GDBP}) @b{actions}
13953 Enter actions for tracepoint #1, one per line.
13954 > collect $regs,$locals,$args
13955 > while-stepping 11
13959 (@value{GDBP}) @b{tstart}
13960 [time passes @dots{}]
13961 (@value{GDBP}) @b{tstop}
13964 @anchor{disconnected tracing}
13965 @cindex disconnected tracing
13966 You can choose to continue running the trace experiment even if
13967 @value{GDBN} disconnects from the target, voluntarily or
13968 involuntarily. For commands such as @code{detach}, the debugger will
13969 ask what you want to do with the trace. But for unexpected
13970 terminations (@value{GDBN} crash, network outage), it would be
13971 unfortunate to lose hard-won trace data, so the variable
13972 @code{disconnected-tracing} lets you decide whether the trace should
13973 continue running without @value{GDBN}.
13976 @item set disconnected-tracing on
13977 @itemx set disconnected-tracing off
13978 @kindex set disconnected-tracing
13979 Choose whether a tracing run should continue to run if @value{GDBN}
13980 has disconnected from the target. Note that @code{detach} or
13981 @code{quit} will ask you directly what to do about a running trace no
13982 matter what this variable's setting, so the variable is mainly useful
13983 for handling unexpected situations, such as loss of the network.
13985 @item show disconnected-tracing
13986 @kindex show disconnected-tracing
13987 Show the current choice for disconnected tracing.
13991 When you reconnect to the target, the trace experiment may or may not
13992 still be running; it might have filled the trace buffer in the
13993 meantime, or stopped for one of the other reasons. If it is running,
13994 it will continue after reconnection.
13996 Upon reconnection, the target will upload information about the
13997 tracepoints in effect. @value{GDBN} will then compare that
13998 information to the set of tracepoints currently defined, and attempt
13999 to match them up, allowing for the possibility that the numbers may
14000 have changed due to creation and deletion in the meantime. If one of
14001 the target's tracepoints does not match any in @value{GDBN}, the
14002 debugger will create a new tracepoint, so that you have a number with
14003 which to specify that tracepoint. This matching-up process is
14004 necessarily heuristic, and it may result in useless tracepoints being
14005 created; you may simply delete them if they are of no use.
14007 @cindex circular trace buffer
14008 If your target agent supports a @dfn{circular trace buffer}, then you
14009 can run a trace experiment indefinitely without filling the trace
14010 buffer; when space runs out, the agent deletes already-collected trace
14011 frames, oldest first, until there is enough room to continue
14012 collecting. This is especially useful if your tracepoints are being
14013 hit too often, and your trace gets terminated prematurely because the
14014 buffer is full. To ask for a circular trace buffer, simply set
14015 @samp{circular-trace-buffer} to on. You can set this at any time,
14016 including during tracing; if the agent can do it, it will change
14017 buffer handling on the fly, otherwise it will not take effect until
14021 @item set circular-trace-buffer on
14022 @itemx set circular-trace-buffer off
14023 @kindex set circular-trace-buffer
14024 Choose whether a tracing run should use a linear or circular buffer
14025 for trace data. A linear buffer will not lose any trace data, but may
14026 fill up prematurely, while a circular buffer will discard old trace
14027 data, but it will have always room for the latest tracepoint hits.
14029 @item show circular-trace-buffer
14030 @kindex show circular-trace-buffer
14031 Show the current choice for the trace buffer. Note that this may not
14032 match the agent's current buffer handling, nor is it guaranteed to
14033 match the setting that might have been in effect during a past run,
14034 for instance if you are looking at frames from a trace file.
14039 @item set trace-buffer-size @var{n}
14040 @itemx set trace-buffer-size unlimited
14041 @kindex set trace-buffer-size
14042 Request that the target use a trace buffer of @var{n} bytes. Not all
14043 targets will honor the request; they may have a compiled-in size for
14044 the trace buffer, or some other limitation. Set to a value of
14045 @code{unlimited} or @code{-1} to let the target use whatever size it
14046 likes. This is also the default.
14048 @item show trace-buffer-size
14049 @kindex show trace-buffer-size
14050 Show the current requested size for the trace buffer. Note that this
14051 will only match the actual size if the target supports size-setting,
14052 and was able to handle the requested size. For instance, if the
14053 target can only change buffer size between runs, this variable will
14054 not reflect the change until the next run starts. Use @code{tstatus}
14055 to get a report of the actual buffer size.
14059 @item set trace-user @var{text}
14060 @kindex set trace-user
14062 @item show trace-user
14063 @kindex show trace-user
14065 @item set trace-notes @var{text}
14066 @kindex set trace-notes
14067 Set the trace run's notes.
14069 @item show trace-notes
14070 @kindex show trace-notes
14071 Show the trace run's notes.
14073 @item set trace-stop-notes @var{text}
14074 @kindex set trace-stop-notes
14075 Set the trace run's stop notes. The handling of the note is as for
14076 @code{tstop} arguments; the set command is convenient way to fix a
14077 stop note that is mistaken or incomplete.
14079 @item show trace-stop-notes
14080 @kindex show trace-stop-notes
14081 Show the trace run's stop notes.
14085 @node Tracepoint Restrictions
14086 @subsection Tracepoint Restrictions
14088 @cindex tracepoint restrictions
14089 There are a number of restrictions on the use of tracepoints. As
14090 described above, tracepoint data gathering occurs on the target
14091 without interaction from @value{GDBN}. Thus the full capabilities of
14092 the debugger are not available during data gathering, and then at data
14093 examination time, you will be limited by only having what was
14094 collected. The following items describe some common problems, but it
14095 is not exhaustive, and you may run into additional difficulties not
14101 Tracepoint expressions are intended to gather objects (lvalues). Thus
14102 the full flexibility of GDB's expression evaluator is not available.
14103 You cannot call functions, cast objects to aggregate types, access
14104 convenience variables or modify values (except by assignment to trace
14105 state variables). Some language features may implicitly call
14106 functions (for instance Objective-C fields with accessors), and therefore
14107 cannot be collected either.
14110 Collection of local variables, either individually or in bulk with
14111 @code{$locals} or @code{$args}, during @code{while-stepping} may
14112 behave erratically. The stepping action may enter a new scope (for
14113 instance by stepping into a function), or the location of the variable
14114 may change (for instance it is loaded into a register). The
14115 tracepoint data recorded uses the location information for the
14116 variables that is correct for the tracepoint location. When the
14117 tracepoint is created, it is not possible, in general, to determine
14118 where the steps of a @code{while-stepping} sequence will advance the
14119 program---particularly if a conditional branch is stepped.
14122 Collection of an incompletely-initialized or partially-destroyed object
14123 may result in something that @value{GDBN} cannot display, or displays
14124 in a misleading way.
14127 When @value{GDBN} displays a pointer to character it automatically
14128 dereferences the pointer to also display characters of the string
14129 being pointed to. However, collecting the pointer during tracing does
14130 not automatically collect the string. You need to explicitly
14131 dereference the pointer and provide size information if you want to
14132 collect not only the pointer, but the memory pointed to. For example,
14133 @code{*ptr@@50} can be used to collect the 50 element array pointed to
14137 It is not possible to collect a complete stack backtrace at a
14138 tracepoint. Instead, you may collect the registers and a few hundred
14139 bytes from the stack pointer with something like @code{*(unsigned char *)$esp@@300}
14140 (adjust to use the name of the actual stack pointer register on your
14141 target architecture, and the amount of stack you wish to capture).
14142 Then the @code{backtrace} command will show a partial backtrace when
14143 using a trace frame. The number of stack frames that can be examined
14144 depends on the sizes of the frames in the collected stack. Note that
14145 if you ask for a block so large that it goes past the bottom of the
14146 stack, the target agent may report an error trying to read from an
14150 If you do not collect registers at a tracepoint, @value{GDBN} can
14151 infer that the value of @code{$pc} must be the same as the address of
14152 the tracepoint and use that when you are looking at a trace frame
14153 for that tracepoint. However, this cannot work if the tracepoint has
14154 multiple locations (for instance if it was set in a function that was
14155 inlined), or if it has a @code{while-stepping} loop. In those cases
14156 @value{GDBN} will warn you that it can't infer @code{$pc}, and default
14161 @node Analyze Collected Data
14162 @section Using the Collected Data
14164 After the tracepoint experiment ends, you use @value{GDBN} commands
14165 for examining the trace data. The basic idea is that each tracepoint
14166 collects a trace @dfn{snapshot} every time it is hit and another
14167 snapshot every time it single-steps. All these snapshots are
14168 consecutively numbered from zero and go into a buffer, and you can
14169 examine them later. The way you examine them is to @dfn{focus} on a
14170 specific trace snapshot. When the remote stub is focused on a trace
14171 snapshot, it will respond to all @value{GDBN} requests for memory and
14172 registers by reading from the buffer which belongs to that snapshot,
14173 rather than from @emph{real} memory or registers of the program being
14174 debugged. This means that @strong{all} @value{GDBN} commands
14175 (@code{print}, @code{info registers}, @code{backtrace}, etc.) will
14176 behave as if we were currently debugging the program state as it was
14177 when the tracepoint occurred. Any requests for data that are not in
14178 the buffer will fail.
14181 * tfind:: How to select a trace snapshot
14182 * tdump:: How to display all data for a snapshot
14183 * save tracepoints:: How to save tracepoints for a future run
14187 @subsection @code{tfind @var{n}}
14190 @cindex select trace snapshot
14191 @cindex find trace snapshot
14192 The basic command for selecting a trace snapshot from the buffer is
14193 @code{tfind @var{n}}, which finds trace snapshot number @var{n},
14194 counting from zero. If no argument @var{n} is given, the next
14195 snapshot is selected.
14197 Here are the various forms of using the @code{tfind} command.
14201 Find the first snapshot in the buffer. This is a synonym for
14202 @code{tfind 0} (since 0 is the number of the first snapshot).
14205 Stop debugging trace snapshots, resume @emph{live} debugging.
14208 Same as @samp{tfind none}.
14211 No argument means find the next trace snapshot or find the first
14212 one if no trace snapshot is selected.
14215 Find the previous trace snapshot before the current one. This permits
14216 retracing earlier steps.
14218 @item tfind tracepoint @var{num}
14219 Find the next snapshot associated with tracepoint @var{num}. Search
14220 proceeds forward from the last examined trace snapshot. If no
14221 argument @var{num} is given, it means find the next snapshot collected
14222 for the same tracepoint as the current snapshot.
14224 @item tfind pc @var{addr}
14225 Find the next snapshot associated with the value @var{addr} of the
14226 program counter. Search proceeds forward from the last examined trace
14227 snapshot. If no argument @var{addr} is given, it means find the next
14228 snapshot with the same value of PC as the current snapshot.
14230 @item tfind outside @var{addr1}, @var{addr2}
14231 Find the next snapshot whose PC is outside the given range of
14232 addresses (exclusive).
14234 @item tfind range @var{addr1}, @var{addr2}
14235 Find the next snapshot whose PC is between @var{addr1} and
14236 @var{addr2} (inclusive).
14238 @item tfind line @r{[}@var{file}:@r{]}@var{n}
14239 Find the next snapshot associated with the source line @var{n}. If
14240 the optional argument @var{file} is given, refer to line @var{n} in
14241 that source file. Search proceeds forward from the last examined
14242 trace snapshot. If no argument @var{n} is given, it means find the
14243 next line other than the one currently being examined; thus saying
14244 @code{tfind line} repeatedly can appear to have the same effect as
14245 stepping from line to line in a @emph{live} debugging session.
14248 The default arguments for the @code{tfind} commands are specifically
14249 designed to make it easy to scan through the trace buffer. For
14250 instance, @code{tfind} with no argument selects the next trace
14251 snapshot, and @code{tfind -} with no argument selects the previous
14252 trace snapshot. So, by giving one @code{tfind} command, and then
14253 simply hitting @key{RET} repeatedly you can examine all the trace
14254 snapshots in order. Or, by saying @code{tfind -} and then hitting
14255 @key{RET} repeatedly you can examine the snapshots in reverse order.
14256 The @code{tfind line} command with no argument selects the snapshot
14257 for the next source line executed. The @code{tfind pc} command with
14258 no argument selects the next snapshot with the same program counter
14259 (PC) as the current frame. The @code{tfind tracepoint} command with
14260 no argument selects the next trace snapshot collected by the same
14261 tracepoint as the current one.
14263 In addition to letting you scan through the trace buffer manually,
14264 these commands make it easy to construct @value{GDBN} scripts that
14265 scan through the trace buffer and print out whatever collected data
14266 you are interested in. Thus, if we want to examine the PC, FP, and SP
14267 registers from each trace frame in the buffer, we can say this:
14270 (@value{GDBP}) @b{tfind start}
14271 (@value{GDBP}) @b{while ($trace_frame != -1)}
14272 > printf "Frame %d, PC = %08X, SP = %08X, FP = %08X\n", \
14273 $trace_frame, $pc, $sp, $fp
14277 Frame 0, PC = 0020DC64, SP = 0030BF3C, FP = 0030BF44
14278 Frame 1, PC = 0020DC6C, SP = 0030BF38, FP = 0030BF44
14279 Frame 2, PC = 0020DC70, SP = 0030BF34, FP = 0030BF44
14280 Frame 3, PC = 0020DC74, SP = 0030BF30, FP = 0030BF44
14281 Frame 4, PC = 0020DC78, SP = 0030BF2C, FP = 0030BF44
14282 Frame 5, PC = 0020DC7C, SP = 0030BF28, FP = 0030BF44
14283 Frame 6, PC = 0020DC80, SP = 0030BF24, FP = 0030BF44
14284 Frame 7, PC = 0020DC84, SP = 0030BF20, FP = 0030BF44
14285 Frame 8, PC = 0020DC88, SP = 0030BF1C, FP = 0030BF44
14286 Frame 9, PC = 0020DC8E, SP = 0030BF18, FP = 0030BF44
14287 Frame 10, PC = 00203F6C, SP = 0030BE3C, FP = 0030BF14
14290 Or, if we want to examine the variable @code{X} at each source line in
14294 (@value{GDBP}) @b{tfind start}
14295 (@value{GDBP}) @b{while ($trace_frame != -1)}
14296 > printf "Frame %d, X == %d\n", $trace_frame, X
14306 @subsection @code{tdump}
14308 @cindex dump all data collected at tracepoint
14309 @cindex tracepoint data, display
14311 This command takes no arguments. It prints all the data collected at
14312 the current trace snapshot.
14315 (@value{GDBP}) @b{trace 444}
14316 (@value{GDBP}) @b{actions}
14317 Enter actions for tracepoint #2, one per line:
14318 > collect $regs, $locals, $args, gdb_long_test
14321 (@value{GDBP}) @b{tstart}
14323 (@value{GDBP}) @b{tfind line 444}
14324 #0 gdb_test (p1=0x11, p2=0x22, p3=0x33, p4=0x44, p5=0x55, p6=0x66)
14326 444 printp( "%s: arguments = 0x%X 0x%X 0x%X 0x%X 0x%X 0x%X\n", )
14328 (@value{GDBP}) @b{tdump}
14329 Data collected at tracepoint 2, trace frame 1:
14330 d0 0xc4aa0085 -995491707
14334 d4 0x71aea3d 119204413
14337 d7 0x380035 3670069
14338 a0 0x19e24a 1696330
14339 a1 0x3000668 50333288
14341 a3 0x322000 3284992
14342 a4 0x3000698 50333336
14343 a5 0x1ad3cc 1758156
14344 fp 0x30bf3c 0x30bf3c
14345 sp 0x30bf34 0x30bf34
14347 pc 0x20b2c8 0x20b2c8
14351 p = 0x20e5b4 "gdb-test"
14358 gdb_long_test = 17 '\021'
14363 @code{tdump} works by scanning the tracepoint's current collection
14364 actions and printing the value of each expression listed. So
14365 @code{tdump} can fail, if after a run, you change the tracepoint's
14366 actions to mention variables that were not collected during the run.
14368 Also, for tracepoints with @code{while-stepping} loops, @code{tdump}
14369 uses the collected value of @code{$pc} to distinguish between trace
14370 frames that were collected at the tracepoint hit, and frames that were
14371 collected while stepping. This allows it to correctly choose whether
14372 to display the basic list of collections, or the collections from the
14373 body of the while-stepping loop. However, if @code{$pc} was not collected,
14374 then @code{tdump} will always attempt to dump using the basic collection
14375 list, and may fail if a while-stepping frame does not include all the
14376 same data that is collected at the tracepoint hit.
14377 @c This is getting pretty arcane, example would be good.
14379 @node save tracepoints
14380 @subsection @code{save tracepoints @var{filename}}
14381 @kindex save tracepoints
14382 @kindex save-tracepoints
14383 @cindex save tracepoints for future sessions
14385 This command saves all current tracepoint definitions together with
14386 their actions and passcounts, into a file @file{@var{filename}}
14387 suitable for use in a later debugging session. To read the saved
14388 tracepoint definitions, use the @code{source} command (@pxref{Command
14389 Files}). The @w{@code{save-tracepoints}} command is a deprecated
14390 alias for @w{@code{save tracepoints}}
14392 @node Tracepoint Variables
14393 @section Convenience Variables for Tracepoints
14394 @cindex tracepoint variables
14395 @cindex convenience variables for tracepoints
14398 @vindex $trace_frame
14399 @item (int) $trace_frame
14400 The current trace snapshot (a.k.a.@: @dfn{frame}) number, or -1 if no
14401 snapshot is selected.
14403 @vindex $tracepoint
14404 @item (int) $tracepoint
14405 The tracepoint for the current trace snapshot.
14407 @vindex $trace_line
14408 @item (int) $trace_line
14409 The line number for the current trace snapshot.
14411 @vindex $trace_file
14412 @item (char []) $trace_file
14413 The source file for the current trace snapshot.
14415 @vindex $trace_func
14416 @item (char []) $trace_func
14417 The name of the function containing @code{$tracepoint}.
14420 Note: @code{$trace_file} is not suitable for use in @code{printf},
14421 use @code{output} instead.
14423 Here's a simple example of using these convenience variables for
14424 stepping through all the trace snapshots and printing some of their
14425 data. Note that these are not the same as trace state variables,
14426 which are managed by the target.
14429 (@value{GDBP}) @b{tfind start}
14431 (@value{GDBP}) @b{while $trace_frame != -1}
14432 > output $trace_file
14433 > printf ", line %d (tracepoint #%d)\n", $trace_line, $tracepoint
14439 @section Using Trace Files
14440 @cindex trace files
14442 In some situations, the target running a trace experiment may no
14443 longer be available; perhaps it crashed, or the hardware was needed
14444 for a different activity. To handle these cases, you can arrange to
14445 dump the trace data into a file, and later use that file as a source
14446 of trace data, via the @code{target tfile} command.
14451 @item tsave [ -r ] @var{filename}
14452 @itemx tsave [-ctf] @var{dirname}
14453 Save the trace data to @var{filename}. By default, this command
14454 assumes that @var{filename} refers to the host filesystem, so if
14455 necessary @value{GDBN} will copy raw trace data up from the target and
14456 then save it. If the target supports it, you can also supply the
14457 optional argument @code{-r} (``remote'') to direct the target to save
14458 the data directly into @var{filename} in its own filesystem, which may be
14459 more efficient if the trace buffer is very large. (Note, however, that
14460 @code{target tfile} can only read from files accessible to the host.)
14461 By default, this command will save trace frame in tfile format.
14462 You can supply the optional argument @code{-ctf} to save data in CTF
14463 format. The @dfn{Common Trace Format} (CTF) is proposed as a trace format
14464 that can be shared by multiple debugging and tracing tools. Please go to
14465 @indicateurl{http://www.efficios.com/ctf} to get more information.
14467 @kindex target tfile
14471 @item target tfile @var{filename}
14472 @itemx target ctf @var{dirname}
14473 Use the file named @var{filename} or directory named @var{dirname} as
14474 a source of trace data. Commands that examine data work as they do with
14475 a live target, but it is not possible to run any new trace experiments.
14476 @code{tstatus} will report the state of the trace run at the moment
14477 the data was saved, as well as the current trace frame you are examining.
14478 Both @var{filename} and @var{dirname} must be on a filesystem accessible to
14482 (@value{GDBP}) target ctf ctf.ctf
14483 (@value{GDBP}) tfind
14484 Found trace frame 0, tracepoint 2
14485 39 ++a; /* set tracepoint 1 here */
14486 (@value{GDBP}) tdump
14487 Data collected at tracepoint 2, trace frame 0:
14491 c = @{"123", "456", "789", "123", "456", "789"@}
14492 d = @{@{@{a = 1, b = 2@}, @{a = 3, b = 4@}@}, @{@{a = 5, b = 6@}, @{a = 7, b = 8@}@}@}
14500 @chapter Debugging Programs That Use Overlays
14503 If your program is too large to fit completely in your target system's
14504 memory, you can sometimes use @dfn{overlays} to work around this
14505 problem. @value{GDBN} provides some support for debugging programs that
14509 * How Overlays Work:: A general explanation of overlays.
14510 * Overlay Commands:: Managing overlays in @value{GDBN}.
14511 * Automatic Overlay Debugging:: @value{GDBN} can find out which overlays are
14512 mapped by asking the inferior.
14513 * Overlay Sample Program:: A sample program using overlays.
14516 @node How Overlays Work
14517 @section How Overlays Work
14518 @cindex mapped overlays
14519 @cindex unmapped overlays
14520 @cindex load address, overlay's
14521 @cindex mapped address
14522 @cindex overlay area
14524 Suppose you have a computer whose instruction address space is only 64
14525 kilobytes long, but which has much more memory which can be accessed by
14526 other means: special instructions, segment registers, or memory
14527 management hardware, for example. Suppose further that you want to
14528 adapt a program which is larger than 64 kilobytes to run on this system.
14530 One solution is to identify modules of your program which are relatively
14531 independent, and need not call each other directly; call these modules
14532 @dfn{overlays}. Separate the overlays from the main program, and place
14533 their machine code in the larger memory. Place your main program in
14534 instruction memory, but leave at least enough space there to hold the
14535 largest overlay as well.
14537 Now, to call a function located in an overlay, you must first copy that
14538 overlay's machine code from the large memory into the space set aside
14539 for it in the instruction memory, and then jump to its entry point
14542 @c NB: In the below the mapped area's size is greater or equal to the
14543 @c size of all overlays. This is intentional to remind the developer
14544 @c that overlays don't necessarily need to be the same size.
14548 Data Instruction Larger
14549 Address Space Address Space Address Space
14550 +-----------+ +-----------+ +-----------+
14552 +-----------+ +-----------+ +-----------+<-- overlay 1
14553 | program | | main | .----| overlay 1 | load address
14554 | variables | | program | | +-----------+
14555 | and heap | | | | | |
14556 +-----------+ | | | +-----------+<-- overlay 2
14557 | | +-----------+ | | | load address
14558 +-----------+ | | | .-| overlay 2 |
14560 mapped --->+-----------+ | | +-----------+
14561 address | | | | | |
14562 | overlay | <-' | | |
14563 | area | <---' +-----------+<-- overlay 3
14564 | | <---. | | load address
14565 +-----------+ `--| overlay 3 |
14572 @anchor{A code overlay}A code overlay
14576 The diagram (@pxref{A code overlay}) shows a system with separate data
14577 and instruction address spaces. To map an overlay, the program copies
14578 its code from the larger address space to the instruction address space.
14579 Since the overlays shown here all use the same mapped address, only one
14580 may be mapped at a time. For a system with a single address space for
14581 data and instructions, the diagram would be similar, except that the
14582 program variables and heap would share an address space with the main
14583 program and the overlay area.
14585 An overlay loaded into instruction memory and ready for use is called a
14586 @dfn{mapped} overlay; its @dfn{mapped address} is its address in the
14587 instruction memory. An overlay not present (or only partially present)
14588 in instruction memory is called @dfn{unmapped}; its @dfn{load address}
14589 is its address in the larger memory. The mapped address is also called
14590 the @dfn{virtual memory address}, or @dfn{VMA}; the load address is also
14591 called the @dfn{load memory address}, or @dfn{LMA}.
14593 Unfortunately, overlays are not a completely transparent way to adapt a
14594 program to limited instruction memory. They introduce a new set of
14595 global constraints you must keep in mind as you design your program:
14600 Before calling or returning to a function in an overlay, your program
14601 must make sure that overlay is actually mapped. Otherwise, the call or
14602 return will transfer control to the right address, but in the wrong
14603 overlay, and your program will probably crash.
14606 If the process of mapping an overlay is expensive on your system, you
14607 will need to choose your overlays carefully to minimize their effect on
14608 your program's performance.
14611 The executable file you load onto your system must contain each
14612 overlay's instructions, appearing at the overlay's load address, not its
14613 mapped address. However, each overlay's instructions must be relocated
14614 and its symbols defined as if the overlay were at its mapped address.
14615 You can use GNU linker scripts to specify different load and relocation
14616 addresses for pieces of your program; see @ref{Overlay Description,,,
14617 ld.info, Using ld: the GNU linker}.
14620 The procedure for loading executable files onto your system must be able
14621 to load their contents into the larger address space as well as the
14622 instruction and data spaces.
14626 The overlay system described above is rather simple, and could be
14627 improved in many ways:
14632 If your system has suitable bank switch registers or memory management
14633 hardware, you could use those facilities to make an overlay's load area
14634 contents simply appear at their mapped address in instruction space.
14635 This would probably be faster than copying the overlay to its mapped
14636 area in the usual way.
14639 If your overlays are small enough, you could set aside more than one
14640 overlay area, and have more than one overlay mapped at a time.
14643 You can use overlays to manage data, as well as instructions. In
14644 general, data overlays are even less transparent to your design than
14645 code overlays: whereas code overlays only require care when you call or
14646 return to functions, data overlays require care every time you access
14647 the data. Also, if you change the contents of a data overlay, you
14648 must copy its contents back out to its load address before you can copy a
14649 different data overlay into the same mapped area.
14654 @node Overlay Commands
14655 @section Overlay Commands
14657 To use @value{GDBN}'s overlay support, each overlay in your program must
14658 correspond to a separate section of the executable file. The section's
14659 virtual memory address and load memory address must be the overlay's
14660 mapped and load addresses. Identifying overlays with sections allows
14661 @value{GDBN} to determine the appropriate address of a function or
14662 variable, depending on whether the overlay is mapped or not.
14664 @value{GDBN}'s overlay commands all start with the word @code{overlay};
14665 you can abbreviate this as @code{ov} or @code{ovly}. The commands are:
14670 Disable @value{GDBN}'s overlay support. When overlay support is
14671 disabled, @value{GDBN} assumes that all functions and variables are
14672 always present at their mapped addresses. By default, @value{GDBN}'s
14673 overlay support is disabled.
14675 @item overlay manual
14676 @cindex manual overlay debugging
14677 Enable @dfn{manual} overlay debugging. In this mode, @value{GDBN}
14678 relies on you to tell it which overlays are mapped, and which are not,
14679 using the @code{overlay map-overlay} and @code{overlay unmap-overlay}
14680 commands described below.
14682 @item overlay map-overlay @var{overlay}
14683 @itemx overlay map @var{overlay}
14684 @cindex map an overlay
14685 Tell @value{GDBN} that @var{overlay} is now mapped; @var{overlay} must
14686 be the name of the object file section containing the overlay. When an
14687 overlay is mapped, @value{GDBN} assumes it can find the overlay's
14688 functions and variables at their mapped addresses. @value{GDBN} assumes
14689 that any other overlays whose mapped ranges overlap that of
14690 @var{overlay} are now unmapped.
14692 @item overlay unmap-overlay @var{overlay}
14693 @itemx overlay unmap @var{overlay}
14694 @cindex unmap an overlay
14695 Tell @value{GDBN} that @var{overlay} is no longer mapped; @var{overlay}
14696 must be the name of the object file section containing the overlay.
14697 When an overlay is unmapped, @value{GDBN} assumes it can find the
14698 overlay's functions and variables at their load addresses.
14701 Enable @dfn{automatic} overlay debugging. In this mode, @value{GDBN}
14702 consults a data structure the overlay manager maintains in the inferior
14703 to see which overlays are mapped. For details, see @ref{Automatic
14704 Overlay Debugging}.
14706 @item overlay load-target
14707 @itemx overlay load
14708 @cindex reloading the overlay table
14709 Re-read the overlay table from the inferior. Normally, @value{GDBN}
14710 re-reads the table @value{GDBN} automatically each time the inferior
14711 stops, so this command should only be necessary if you have changed the
14712 overlay mapping yourself using @value{GDBN}. This command is only
14713 useful when using automatic overlay debugging.
14715 @item overlay list-overlays
14716 @itemx overlay list
14717 @cindex listing mapped overlays
14718 Display a list of the overlays currently mapped, along with their mapped
14719 addresses, load addresses, and sizes.
14723 Normally, when @value{GDBN} prints a code address, it includes the name
14724 of the function the address falls in:
14727 (@value{GDBP}) print main
14728 $3 = @{int ()@} 0x11a0 <main>
14731 When overlay debugging is enabled, @value{GDBN} recognizes code in
14732 unmapped overlays, and prints the names of unmapped functions with
14733 asterisks around them. For example, if @code{foo} is a function in an
14734 unmapped overlay, @value{GDBN} prints it this way:
14737 (@value{GDBP}) overlay list
14738 No sections are mapped.
14739 (@value{GDBP}) print foo
14740 $5 = @{int (int)@} 0x100000 <*foo*>
14743 When @code{foo}'s overlay is mapped, @value{GDBN} prints the function's
14747 (@value{GDBP}) overlay list
14748 Section .ov.foo.text, loaded at 0x100000 - 0x100034,
14749 mapped at 0x1016 - 0x104a
14750 (@value{GDBP}) print foo
14751 $6 = @{int (int)@} 0x1016 <foo>
14754 When overlay debugging is enabled, @value{GDBN} can find the correct
14755 address for functions and variables in an overlay, whether or not the
14756 overlay is mapped. This allows most @value{GDBN} commands, like
14757 @code{break} and @code{disassemble}, to work normally, even on unmapped
14758 code. However, @value{GDBN}'s breakpoint support has some limitations:
14762 @cindex breakpoints in overlays
14763 @cindex overlays, setting breakpoints in
14764 You can set breakpoints in functions in unmapped overlays, as long as
14765 @value{GDBN} can write to the overlay at its load address.
14767 @value{GDBN} can not set hardware or simulator-based breakpoints in
14768 unmapped overlays. However, if you set a breakpoint at the end of your
14769 overlay manager (and tell @value{GDBN} which overlays are now mapped, if
14770 you are using manual overlay management), @value{GDBN} will re-set its
14771 breakpoints properly.
14775 @node Automatic Overlay Debugging
14776 @section Automatic Overlay Debugging
14777 @cindex automatic overlay debugging
14779 @value{GDBN} can automatically track which overlays are mapped and which
14780 are not, given some simple co-operation from the overlay manager in the
14781 inferior. If you enable automatic overlay debugging with the
14782 @code{overlay auto} command (@pxref{Overlay Commands}), @value{GDBN}
14783 looks in the inferior's memory for certain variables describing the
14784 current state of the overlays.
14786 Here are the variables your overlay manager must define to support
14787 @value{GDBN}'s automatic overlay debugging:
14791 @item @code{_ovly_table}:
14792 This variable must be an array of the following structures:
14797 /* The overlay's mapped address. */
14800 /* The size of the overlay, in bytes. */
14801 unsigned long size;
14803 /* The overlay's load address. */
14806 /* Non-zero if the overlay is currently mapped;
14808 unsigned long mapped;
14812 @item @code{_novlys}:
14813 This variable must be a four-byte signed integer, holding the total
14814 number of elements in @code{_ovly_table}.
14818 To decide whether a particular overlay is mapped or not, @value{GDBN}
14819 looks for an entry in @w{@code{_ovly_table}} whose @code{vma} and
14820 @code{lma} members equal the VMA and LMA of the overlay's section in the
14821 executable file. When @value{GDBN} finds a matching entry, it consults
14822 the entry's @code{mapped} member to determine whether the overlay is
14825 In addition, your overlay manager may define a function called
14826 @code{_ovly_debug_event}. If this function is defined, @value{GDBN}
14827 will silently set a breakpoint there. If the overlay manager then
14828 calls this function whenever it has changed the overlay table, this
14829 will enable @value{GDBN} to accurately keep track of which overlays
14830 are in program memory, and update any breakpoints that may be set
14831 in overlays. This will allow breakpoints to work even if the
14832 overlays are kept in ROM or other non-writable memory while they
14833 are not being executed.
14835 @node Overlay Sample Program
14836 @section Overlay Sample Program
14837 @cindex overlay example program
14839 When linking a program which uses overlays, you must place the overlays
14840 at their load addresses, while relocating them to run at their mapped
14841 addresses. To do this, you must write a linker script (@pxref{Overlay
14842 Description,,, ld.info, Using ld: the GNU linker}). Unfortunately,
14843 since linker scripts are specific to a particular host system, target
14844 architecture, and target memory layout, this manual cannot provide
14845 portable sample code demonstrating @value{GDBN}'s overlay support.
14847 However, the @value{GDBN} source distribution does contain an overlaid
14848 program, with linker scripts for a few systems, as part of its test
14849 suite. The program consists of the following files from
14850 @file{gdb/testsuite/gdb.base}:
14854 The main program file.
14856 A simple overlay manager, used by @file{overlays.c}.
14861 Overlay modules, loaded and used by @file{overlays.c}.
14864 Linker scripts for linking the test program on the @code{d10v-elf}
14865 and @code{m32r-elf} targets.
14868 You can build the test program using the @code{d10v-elf} GCC
14869 cross-compiler like this:
14872 $ d10v-elf-gcc -g -c overlays.c
14873 $ d10v-elf-gcc -g -c ovlymgr.c
14874 $ d10v-elf-gcc -g -c foo.c
14875 $ d10v-elf-gcc -g -c bar.c
14876 $ d10v-elf-gcc -g -c baz.c
14877 $ d10v-elf-gcc -g -c grbx.c
14878 $ d10v-elf-gcc -g overlays.o ovlymgr.o foo.o bar.o \
14879 baz.o grbx.o -Wl,-Td10v.ld -o overlays
14882 The build process is identical for any other architecture, except that
14883 you must substitute the appropriate compiler and linker script for the
14884 target system for @code{d10v-elf-gcc} and @code{d10v.ld}.
14888 @chapter Using @value{GDBN} with Different Languages
14891 Although programming languages generally have common aspects, they are
14892 rarely expressed in the same manner. For instance, in ANSI C,
14893 dereferencing a pointer @code{p} is accomplished by @code{*p}, but in
14894 Modula-2, it is accomplished by @code{p^}. Values can also be
14895 represented (and displayed) differently. Hex numbers in C appear as
14896 @samp{0x1ae}, while in Modula-2 they appear as @samp{1AEH}.
14898 @cindex working language
14899 Language-specific information is built into @value{GDBN} for some languages,
14900 allowing you to express operations like the above in your program's
14901 native language, and allowing @value{GDBN} to output values in a manner
14902 consistent with the syntax of your program's native language. The
14903 language you use to build expressions is called the @dfn{working
14907 * Setting:: Switching between source languages
14908 * Show:: Displaying the language
14909 * Checks:: Type and range checks
14910 * Supported Languages:: Supported languages
14911 * Unsupported Languages:: Unsupported languages
14915 @section Switching Between Source Languages
14917 There are two ways to control the working language---either have @value{GDBN}
14918 set it automatically, or select it manually yourself. You can use the
14919 @code{set language} command for either purpose. On startup, @value{GDBN}
14920 defaults to setting the language automatically. The working language is
14921 used to determine how expressions you type are interpreted, how values
14924 In addition to the working language, every source file that
14925 @value{GDBN} knows about has its own working language. For some object
14926 file formats, the compiler might indicate which language a particular
14927 source file is in. However, most of the time @value{GDBN} infers the
14928 language from the name of the file. The language of a source file
14929 controls whether C@t{++} names are demangled---this way @code{backtrace} can
14930 show each frame appropriately for its own language. There is no way to
14931 set the language of a source file from within @value{GDBN}, but you can
14932 set the language associated with a filename extension. @xref{Show, ,
14933 Displaying the Language}.
14935 This is most commonly a problem when you use a program, such
14936 as @code{cfront} or @code{f2c}, that generates C but is written in
14937 another language. In that case, make the
14938 program use @code{#line} directives in its C output; that way
14939 @value{GDBN} will know the correct language of the source code of the original
14940 program, and will display that source code, not the generated C code.
14943 * Filenames:: Filename extensions and languages.
14944 * Manually:: Setting the working language manually
14945 * Automatically:: Having @value{GDBN} infer the source language
14949 @subsection List of Filename Extensions and Languages
14951 If a source file name ends in one of the following extensions, then
14952 @value{GDBN} infers that its language is the one indicated.
14970 C@t{++} source file
14976 Objective-C source file
14980 Fortran source file
14983 Modula-2 source file
14987 Assembler source file. This actually behaves almost like C, but
14988 @value{GDBN} does not skip over function prologues when stepping.
14991 In addition, you may set the language associated with a filename
14992 extension. @xref{Show, , Displaying the Language}.
14995 @subsection Setting the Working Language
14997 If you allow @value{GDBN} to set the language automatically,
14998 expressions are interpreted the same way in your debugging session and
15001 @kindex set language
15002 If you wish, you may set the language manually. To do this, issue the
15003 command @samp{set language @var{lang}}, where @var{lang} is the name of
15004 a language, such as
15005 @code{c} or @code{modula-2}.
15006 For a list of the supported languages, type @samp{set language}.
15008 Setting the language manually prevents @value{GDBN} from updating the working
15009 language automatically. This can lead to confusion if you try
15010 to debug a program when the working language is not the same as the
15011 source language, when an expression is acceptable to both
15012 languages---but means different things. For instance, if the current
15013 source file were written in C, and @value{GDBN} was parsing Modula-2, a
15021 might not have the effect you intended. In C, this means to add
15022 @code{b} and @code{c} and place the result in @code{a}. The result
15023 printed would be the value of @code{a}. In Modula-2, this means to compare
15024 @code{a} to the result of @code{b+c}, yielding a @code{BOOLEAN} value.
15026 @node Automatically
15027 @subsection Having @value{GDBN} Infer the Source Language
15029 To have @value{GDBN} set the working language automatically, use
15030 @samp{set language local} or @samp{set language auto}. @value{GDBN}
15031 then infers the working language. That is, when your program stops in a
15032 frame (usually by encountering a breakpoint), @value{GDBN} sets the
15033 working language to the language recorded for the function in that
15034 frame. If the language for a frame is unknown (that is, if the function
15035 or block corresponding to the frame was defined in a source file that
15036 does not have a recognized extension), the current working language is
15037 not changed, and @value{GDBN} issues a warning.
15039 This may not seem necessary for most programs, which are written
15040 entirely in one source language. However, program modules and libraries
15041 written in one source language can be used by a main program written in
15042 a different source language. Using @samp{set language auto} in this
15043 case frees you from having to set the working language manually.
15046 @section Displaying the Language
15048 The following commands help you find out which language is the
15049 working language, and also what language source files were written in.
15052 @item show language
15053 @anchor{show language}
15054 @kindex show language
15055 Display the current working language. This is the
15056 language you can use with commands such as @code{print} to
15057 build and compute expressions that may involve variables in your program.
15060 @kindex info frame@r{, show the source language}
15061 Display the source language for this frame. This language becomes the
15062 working language if you use an identifier from this frame.
15063 @xref{Frame Info, ,Information about a Frame}, to identify the other
15064 information listed here.
15067 @kindex info source@r{, show the source language}
15068 Display the source language of this source file.
15069 @xref{Symbols, ,Examining the Symbol Table}, to identify the other
15070 information listed here.
15073 In unusual circumstances, you may have source files with extensions
15074 not in the standard list. You can then set the extension associated
15075 with a language explicitly:
15078 @item set extension-language @var{ext} @var{language}
15079 @kindex set extension-language
15080 Tell @value{GDBN} that source files with extension @var{ext} are to be
15081 assumed as written in the source language @var{language}.
15083 @item info extensions
15084 @kindex info extensions
15085 List all the filename extensions and the associated languages.
15089 @section Type and Range Checking
15091 Some languages are designed to guard you against making seemingly common
15092 errors through a series of compile- and run-time checks. These include
15093 checking the type of arguments to functions and operators and making
15094 sure mathematical overflows are caught at run time. Checks such as
15095 these help to ensure a program's correctness once it has been compiled
15096 by eliminating type mismatches and providing active checks for range
15097 errors when your program is running.
15099 By default @value{GDBN} checks for these errors according to the
15100 rules of the current source language. Although @value{GDBN} does not check
15101 the statements in your program, it can check expressions entered directly
15102 into @value{GDBN} for evaluation via the @code{print} command, for example.
15105 * Type Checking:: An overview of type checking
15106 * Range Checking:: An overview of range checking
15109 @cindex type checking
15110 @cindex checks, type
15111 @node Type Checking
15112 @subsection An Overview of Type Checking
15114 Some languages, such as C and C@t{++}, are strongly typed, meaning that the
15115 arguments to operators and functions have to be of the correct type,
15116 otherwise an error occurs. These checks prevent type mismatch
15117 errors from ever causing any run-time problems. For example,
15120 int klass::my_method(char *b) @{ return b ? 1 : 2; @}
15122 (@value{GDBP}) print obj.my_method (0)
15125 (@value{GDBP}) print obj.my_method (0x1234)
15126 Cannot resolve method klass::my_method to any overloaded instance
15129 The second example fails because in C@t{++} the integer constant
15130 @samp{0x1234} is not type-compatible with the pointer parameter type.
15132 For the expressions you use in @value{GDBN} commands, you can tell
15133 @value{GDBN} to not enforce strict type checking or
15134 to treat any mismatches as errors and abandon the expression;
15135 When type checking is disabled, @value{GDBN} successfully evaluates
15136 expressions like the second example above.
15138 Even if type checking is off, there may be other reasons
15139 related to type that prevent @value{GDBN} from evaluating an expression.
15140 For instance, @value{GDBN} does not know how to add an @code{int} and
15141 a @code{struct foo}. These particular type errors have nothing to do
15142 with the language in use and usually arise from expressions which make
15143 little sense to evaluate anyway.
15145 @value{GDBN} provides some additional commands for controlling type checking:
15147 @kindex set check type
15148 @kindex show check type
15150 @item set check type on
15151 @itemx set check type off
15152 Set strict type checking on or off. If any type mismatches occur in
15153 evaluating an expression while type checking is on, @value{GDBN} prints a
15154 message and aborts evaluation of the expression.
15156 @item show check type
15157 Show the current setting of type checking and whether @value{GDBN}
15158 is enforcing strict type checking rules.
15161 @cindex range checking
15162 @cindex checks, range
15163 @node Range Checking
15164 @subsection An Overview of Range Checking
15166 In some languages (such as Modula-2), it is an error to exceed the
15167 bounds of a type; this is enforced with run-time checks. Such range
15168 checking is meant to ensure program correctness by making sure
15169 computations do not overflow, or indices on an array element access do
15170 not exceed the bounds of the array.
15172 For expressions you use in @value{GDBN} commands, you can tell
15173 @value{GDBN} to treat range errors in one of three ways: ignore them,
15174 always treat them as errors and abandon the expression, or issue
15175 warnings but evaluate the expression anyway.
15177 A range error can result from numerical overflow, from exceeding an
15178 array index bound, or when you type a constant that is not a member
15179 of any type. Some languages, however, do not treat overflows as an
15180 error. In many implementations of C, mathematical overflow causes the
15181 result to ``wrap around'' to lower values---for example, if @var{m} is
15182 the largest integer value, and @var{s} is the smallest, then
15185 @var{m} + 1 @result{} @var{s}
15188 This, too, is specific to individual languages, and in some cases
15189 specific to individual compilers or machines. @xref{Supported Languages, ,
15190 Supported Languages}, for further details on specific languages.
15192 @value{GDBN} provides some additional commands for controlling the range checker:
15194 @kindex set check range
15195 @kindex show check range
15197 @item set check range auto
15198 Set range checking on or off based on the current working language.
15199 @xref{Supported Languages, ,Supported Languages}, for the default settings for
15202 @item set check range on
15203 @itemx set check range off
15204 Set range checking on or off, overriding the default setting for the
15205 current working language. A warning is issued if the setting does not
15206 match the language default. If a range error occurs and range checking is on,
15207 then a message is printed and evaluation of the expression is aborted.
15209 @item set check range warn
15210 Output messages when the @value{GDBN} range checker detects a range error,
15211 but attempt to evaluate the expression anyway. Evaluating the
15212 expression may still be impossible for other reasons, such as accessing
15213 memory that the process does not own (a typical example from many Unix
15217 Show the current setting of the range checker, and whether or not it is
15218 being set automatically by @value{GDBN}.
15221 @node Supported Languages
15222 @section Supported Languages
15224 @value{GDBN} supports C, C@t{++}, D, Go, Objective-C, Fortran,
15225 OpenCL C, Pascal, Rust, assembly, Modula-2, and Ada.
15226 @c This is false ...
15227 Some @value{GDBN} features may be used in expressions regardless of the
15228 language you use: the @value{GDBN} @code{@@} and @code{::} operators,
15229 and the @samp{@{type@}addr} construct (@pxref{Expressions,
15230 ,Expressions}) can be used with the constructs of any supported
15233 The following sections detail to what degree each source language is
15234 supported by @value{GDBN}. These sections are not meant to be language
15235 tutorials or references, but serve only as a reference guide to what the
15236 @value{GDBN} expression parser accepts, and what input and output
15237 formats should look like for different languages. There are many good
15238 books written on each of these languages; please look to these for a
15239 language reference or tutorial.
15242 * C:: C and C@t{++}
15245 * Objective-C:: Objective-C
15246 * OpenCL C:: OpenCL C
15247 * Fortran:: Fortran
15250 * Modula-2:: Modula-2
15255 @subsection C and C@t{++}
15257 @cindex C and C@t{++}
15258 @cindex expressions in C or C@t{++}
15260 Since C and C@t{++} are so closely related, many features of @value{GDBN} apply
15261 to both languages. Whenever this is the case, we discuss those languages
15265 @cindex @code{g++}, @sc{gnu} C@t{++} compiler
15266 @cindex @sc{gnu} C@t{++}
15267 The C@t{++} debugging facilities are jointly implemented by the C@t{++}
15268 compiler and @value{GDBN}. Therefore, to debug your C@t{++} code
15269 effectively, you must compile your C@t{++} programs with a supported
15270 C@t{++} compiler, such as @sc{gnu} @code{g++}, or the HP ANSI C@t{++}
15271 compiler (@code{aCC}).
15274 * C Operators:: C and C@t{++} operators
15275 * C Constants:: C and C@t{++} constants
15276 * C Plus Plus Expressions:: C@t{++} expressions
15277 * C Defaults:: Default settings for C and C@t{++}
15278 * C Checks:: C and C@t{++} type and range checks
15279 * Debugging C:: @value{GDBN} and C
15280 * Debugging C Plus Plus:: @value{GDBN} features for C@t{++}
15281 * Decimal Floating Point:: Numbers in Decimal Floating Point format
15285 @subsubsection C and C@t{++} Operators
15287 @cindex C and C@t{++} operators
15289 Operators must be defined on values of specific types. For instance,
15290 @code{+} is defined on numbers, but not on structures. Operators are
15291 often defined on groups of types.
15293 For the purposes of C and C@t{++}, the following definitions hold:
15298 @emph{Integral types} include @code{int} with any of its storage-class
15299 specifiers; @code{char}; @code{enum}; and, for C@t{++}, @code{bool}.
15302 @emph{Floating-point types} include @code{float}, @code{double}, and
15303 @code{long double} (if supported by the target platform).
15306 @emph{Pointer types} include all types defined as @code{(@var{type} *)}.
15309 @emph{Scalar types} include all of the above.
15314 The following operators are supported. They are listed here
15315 in order of increasing precedence:
15319 The comma or sequencing operator. Expressions in a comma-separated list
15320 are evaluated from left to right, with the result of the entire
15321 expression being the last expression evaluated.
15324 Assignment. The value of an assignment expression is the value
15325 assigned. Defined on scalar types.
15328 Used in an expression of the form @w{@code{@var{a} @var{op}= @var{b}}},
15329 and translated to @w{@code{@var{a} = @var{a op b}}}.
15330 @w{@code{@var{op}=}} and @code{=} have the same precedence. The operator
15331 @var{op} is any one of the operators @code{|}, @code{^}, @code{&},
15332 @code{<<}, @code{>>}, @code{+}, @code{-}, @code{*}, @code{/}, @code{%}.
15335 The ternary operator. @code{@var{a} ? @var{b} : @var{c}} can be thought
15336 of as: if @var{a} then @var{b} else @var{c}. The argument @var{a}
15337 should be of an integral type.
15340 Logical @sc{or}. Defined on integral types.
15343 Logical @sc{and}. Defined on integral types.
15346 Bitwise @sc{or}. Defined on integral types.
15349 Bitwise exclusive-@sc{or}. Defined on integral types.
15352 Bitwise @sc{and}. Defined on integral types.
15355 Equality and inequality. Defined on scalar types. The value of these
15356 expressions is 0 for false and non-zero for true.
15358 @item <@r{, }>@r{, }<=@r{, }>=
15359 Less than, greater than, less than or equal, greater than or equal.
15360 Defined on scalar types. The value of these expressions is 0 for false
15361 and non-zero for true.
15364 left shift, and right shift. Defined on integral types.
15367 The @value{GDBN} ``artificial array'' operator (@pxref{Expressions, ,Expressions}).
15370 Addition and subtraction. Defined on integral types, floating-point types and
15373 @item *@r{, }/@r{, }%
15374 Multiplication, division, and modulus. Multiplication and division are
15375 defined on integral and floating-point types. Modulus is defined on
15379 Increment and decrement. When appearing before a variable, the
15380 operation is performed before the variable is used in an expression;
15381 when appearing after it, the variable's value is used before the
15382 operation takes place.
15385 Pointer dereferencing. Defined on pointer types. Same precedence as
15389 Address operator. Defined on variables. Same precedence as @code{++}.
15391 For debugging C@t{++}, @value{GDBN} implements a use of @samp{&} beyond what is
15392 allowed in the C@t{++} language itself: you can use @samp{&(&@var{ref})}
15393 to examine the address
15394 where a C@t{++} reference variable (declared with @samp{&@var{ref}}) is
15398 Negative. Defined on integral and floating-point types. Same
15399 precedence as @code{++}.
15402 Logical negation. Defined on integral types. Same precedence as
15406 Bitwise complement operator. Defined on integral types. Same precedence as
15411 Structure member, and pointer-to-structure member. For convenience,
15412 @value{GDBN} regards the two as equivalent, choosing whether to dereference a
15413 pointer based on the stored type information.
15414 Defined on @code{struct} and @code{union} data.
15417 Dereferences of pointers to members.
15420 Array indexing. @code{@var{a}[@var{i}]} is defined as
15421 @code{*(@var{a}+@var{i})}. Same precedence as @code{->}.
15424 Function parameter list. Same precedence as @code{->}.
15427 C@t{++} scope resolution operator. Defined on @code{struct}, @code{union},
15428 and @code{class} types.
15431 Doubled colons also represent the @value{GDBN} scope operator
15432 (@pxref{Expressions, ,Expressions}). Same precedence as @code{::},
15436 If an operator is redefined in the user code, @value{GDBN} usually
15437 attempts to invoke the redefined version instead of using the operator's
15438 predefined meaning.
15441 @subsubsection C and C@t{++} Constants
15443 @cindex C and C@t{++} constants
15445 @value{GDBN} allows you to express the constants of C and C@t{++} in the
15450 Integer constants are a sequence of digits. Octal constants are
15451 specified by a leading @samp{0} (i.e.@: zero), and hexadecimal constants
15452 by a leading @samp{0x} or @samp{0X}. Constants may also end with a letter
15453 @samp{l}, specifying that the constant should be treated as a
15457 Floating point constants are a sequence of digits, followed by a decimal
15458 point, followed by a sequence of digits, and optionally followed by an
15459 exponent. An exponent is of the form:
15460 @samp{@w{e@r{[[}+@r{]|}-@r{]}@var{nnn}}}, where @var{nnn} is another
15461 sequence of digits. The @samp{+} is optional for positive exponents.
15462 A floating-point constant may also end with a letter @samp{f} or
15463 @samp{F}, specifying that the constant should be treated as being of
15464 the @code{float} (as opposed to the default @code{double}) type; or with
15465 a letter @samp{l} or @samp{L}, which specifies a @code{long double}
15469 Enumerated constants consist of enumerated identifiers, or their
15470 integral equivalents.
15473 Character constants are a single character surrounded by single quotes
15474 (@code{'}), or a number---the ordinal value of the corresponding character
15475 (usually its @sc{ascii} value). Within quotes, the single character may
15476 be represented by a letter or by @dfn{escape sequences}, which are of
15477 the form @samp{\@var{nnn}}, where @var{nnn} is the octal representation
15478 of the character's ordinal value; or of the form @samp{\@var{x}}, where
15479 @samp{@var{x}} is a predefined special character---for example,
15480 @samp{\n} for newline.
15482 Wide character constants can be written by prefixing a character
15483 constant with @samp{L}, as in C. For example, @samp{L'x'} is the wide
15484 form of @samp{x}. The target wide character set is used when
15485 computing the value of this constant (@pxref{Character Sets}).
15488 String constants are a sequence of character constants surrounded by
15489 double quotes (@code{"}). Any valid character constant (as described
15490 above) may appear. Double quotes within the string must be preceded by
15491 a backslash, so for instance @samp{"a\"b'c"} is a string of five
15494 Wide string constants can be written by prefixing a string constant
15495 with @samp{L}, as in C. The target wide character set is used when
15496 computing the value of this constant (@pxref{Character Sets}).
15499 Pointer constants are an integral value. You can also write pointers
15500 to constants using the C operator @samp{&}.
15503 Array constants are comma-separated lists surrounded by braces @samp{@{}
15504 and @samp{@}}; for example, @samp{@{1,2,3@}} is a three-element array of
15505 integers, @samp{@{@{1,2@}, @{3,4@}, @{5,6@}@}} is a three-by-two array,
15506 and @samp{@{&"hi", &"there", &"fred"@}} is a three-element array of pointers.
15509 @node C Plus Plus Expressions
15510 @subsubsection C@t{++} Expressions
15512 @cindex expressions in C@t{++}
15513 @value{GDBN} expression handling can interpret most C@t{++} expressions.
15515 @cindex debugging C@t{++} programs
15516 @cindex C@t{++} compilers
15517 @cindex debug formats and C@t{++}
15518 @cindex @value{NGCC} and C@t{++}
15520 @emph{Warning:} @value{GDBN} can only debug C@t{++} code if you use
15521 the proper compiler and the proper debug format. Currently,
15522 @value{GDBN} works best when debugging C@t{++} code that is compiled
15523 with the most recent version of @value{NGCC} possible. The DWARF
15524 debugging format is preferred; @value{NGCC} defaults to this on most
15525 popular platforms. Other compilers and/or debug formats are likely to
15526 work badly or not at all when using @value{GDBN} to debug C@t{++}
15527 code. @xref{Compilation}.
15532 @cindex member functions
15534 Member function calls are allowed; you can use expressions like
15537 count = aml->GetOriginal(x, y)
15540 @vindex this@r{, inside C@t{++} member functions}
15541 @cindex namespace in C@t{++}
15543 While a member function is active (in the selected stack frame), your
15544 expressions have the same namespace available as the member function;
15545 that is, @value{GDBN} allows implicit references to the class instance
15546 pointer @code{this} following the same rules as C@t{++}. @code{using}
15547 declarations in the current scope are also respected by @value{GDBN}.
15549 @cindex call overloaded functions
15550 @cindex overloaded functions, calling
15551 @cindex type conversions in C@t{++}
15553 You can call overloaded functions; @value{GDBN} resolves the function
15554 call to the right definition, with some restrictions. @value{GDBN} does not
15555 perform overload resolution involving user-defined type conversions,
15556 calls to constructors, or instantiations of templates that do not exist
15557 in the program. It also cannot handle ellipsis argument lists or
15560 It does perform integral conversions and promotions, floating-point
15561 promotions, arithmetic conversions, pointer conversions, conversions of
15562 class objects to base classes, and standard conversions such as those of
15563 functions or arrays to pointers; it requires an exact match on the
15564 number of function arguments.
15566 Overload resolution is always performed, unless you have specified
15567 @code{set overload-resolution off}. @xref{Debugging C Plus Plus,
15568 ,@value{GDBN} Features for C@t{++}}.
15570 You must specify @code{set overload-resolution off} in order to use an
15571 explicit function signature to call an overloaded function, as in
15573 p 'foo(char,int)'('x', 13)
15576 The @value{GDBN} command-completion facility can simplify this;
15577 see @ref{Completion, ,Command Completion}.
15579 @cindex reference declarations
15581 @value{GDBN} understands variables declared as C@t{++} lvalue or rvalue
15582 references; you can use them in expressions just as you do in C@t{++}
15583 source---they are automatically dereferenced.
15585 In the parameter list shown when @value{GDBN} displays a frame, the values of
15586 reference variables are not displayed (unlike other variables); this
15587 avoids clutter, since references are often used for large structures.
15588 The @emph{address} of a reference variable is always shown, unless
15589 you have specified @samp{set print address off}.
15592 @value{GDBN} supports the C@t{++} name resolution operator @code{::}---your
15593 expressions can use it just as expressions in your program do. Since
15594 one scope may be defined in another, you can use @code{::} repeatedly if
15595 necessary, for example in an expression like
15596 @samp{@var{scope1}::@var{scope2}::@var{name}}. @value{GDBN} also allows
15597 resolving name scope by reference to source files, in both C and C@t{++}
15598 debugging (@pxref{Variables, ,Program Variables}).
15601 @value{GDBN} performs argument-dependent lookup, following the C@t{++}
15606 @subsubsection C and C@t{++} Defaults
15608 @cindex C and C@t{++} defaults
15610 If you allow @value{GDBN} to set range checking automatically, it
15611 defaults to @code{off} whenever the working language changes to
15612 C or C@t{++}. This happens regardless of whether you or @value{GDBN}
15613 selects the working language.
15615 If you allow @value{GDBN} to set the language automatically, it
15616 recognizes source files whose names end with @file{.c}, @file{.C}, or
15617 @file{.cc}, etc, and when @value{GDBN} enters code compiled from one of
15618 these files, it sets the working language to C or C@t{++}.
15619 @xref{Automatically, ,Having @value{GDBN} Infer the Source Language},
15620 for further details.
15623 @subsubsection C and C@t{++} Type and Range Checks
15625 @cindex C and C@t{++} checks
15627 By default, when @value{GDBN} parses C or C@t{++} expressions, strict type
15628 checking is used. However, if you turn type checking off, @value{GDBN}
15629 will allow certain non-standard conversions, such as promoting integer
15630 constants to pointers.
15632 Range checking, if turned on, is done on mathematical operations. Array
15633 indices are not checked, since they are often used to index a pointer
15634 that is not itself an array.
15637 @subsubsection @value{GDBN} and C
15639 The @code{set print union} and @code{show print union} commands apply to
15640 the @code{union} type. When set to @samp{on}, any @code{union} that is
15641 inside a @code{struct} or @code{class} is also printed. Otherwise, it
15642 appears as @samp{@{...@}}.
15644 The @code{@@} operator aids in the debugging of dynamic arrays, formed
15645 with pointers and a memory allocation function. @xref{Expressions,
15648 @node Debugging C Plus Plus
15649 @subsubsection @value{GDBN} Features for C@t{++}
15651 @cindex commands for C@t{++}
15653 Some @value{GDBN} commands are particularly useful with C@t{++}, and some are
15654 designed specifically for use with C@t{++}. Here is a summary:
15657 @cindex break in overloaded functions
15658 @item @r{breakpoint menus}
15659 When you want a breakpoint in a function whose name is overloaded,
15660 @value{GDBN} has the capability to display a menu of possible breakpoint
15661 locations to help you specify which function definition you want.
15662 @xref{Ambiguous Expressions,,Ambiguous Expressions}.
15664 @cindex overloading in C@t{++}
15665 @item rbreak @var{regex}
15666 Setting breakpoints using regular expressions is helpful for setting
15667 breakpoints on overloaded functions that are not members of any special
15669 @xref{Set Breaks, ,Setting Breakpoints}.
15671 @cindex C@t{++} exception handling
15673 @itemx catch rethrow
15675 Debug C@t{++} exception handling using these commands. @xref{Set
15676 Catchpoints, , Setting Catchpoints}.
15678 @cindex inheritance
15679 @item ptype @var{typename}
15680 Print inheritance relationships as well as other information for type
15682 @xref{Symbols, ,Examining the Symbol Table}.
15684 @item info vtbl @var{expression}.
15685 The @code{info vtbl} command can be used to display the virtual
15686 method tables of the object computed by @var{expression}. This shows
15687 one entry per virtual table; there may be multiple virtual tables when
15688 multiple inheritance is in use.
15690 @cindex C@t{++} demangling
15691 @item demangle @var{name}
15692 Demangle @var{name}.
15693 @xref{Symbols}, for a more complete description of the @code{demangle} command.
15695 @cindex C@t{++} symbol display
15696 @item set print demangle
15697 @itemx show print demangle
15698 @itemx set print asm-demangle
15699 @itemx show print asm-demangle
15700 Control whether C@t{++} symbols display in their source form, both when
15701 displaying code as C@t{++} source and when displaying disassemblies.
15702 @xref{Print Settings, ,Print Settings}.
15704 @item set print object
15705 @itemx show print object
15706 Choose whether to print derived (actual) or declared types of objects.
15707 @xref{Print Settings, ,Print Settings}.
15709 @item set print vtbl
15710 @itemx show print vtbl
15711 Control the format for printing virtual function tables.
15712 @xref{Print Settings, ,Print Settings}.
15713 (The @code{vtbl} commands do not work on programs compiled with the HP
15714 ANSI C@t{++} compiler (@code{aCC}).)
15716 @kindex set overload-resolution
15717 @cindex overloaded functions, overload resolution
15718 @item set overload-resolution on
15719 Enable overload resolution for C@t{++} expression evaluation. The default
15720 is on. For overloaded functions, @value{GDBN} evaluates the arguments
15721 and searches for a function whose signature matches the argument types,
15722 using the standard C@t{++} conversion rules (see @ref{C Plus Plus
15723 Expressions, ,C@t{++} Expressions}, for details).
15724 If it cannot find a match, it emits a message.
15726 @item set overload-resolution off
15727 Disable overload resolution for C@t{++} expression evaluation. For
15728 overloaded functions that are not class member functions, @value{GDBN}
15729 chooses the first function of the specified name that it finds in the
15730 symbol table, whether or not its arguments are of the correct type. For
15731 overloaded functions that are class member functions, @value{GDBN}
15732 searches for a function whose signature @emph{exactly} matches the
15735 @kindex show overload-resolution
15736 @item show overload-resolution
15737 Show the current setting of overload resolution.
15739 @item @r{Overloaded symbol names}
15740 You can specify a particular definition of an overloaded symbol, using
15741 the same notation that is used to declare such symbols in C@t{++}: type
15742 @code{@var{symbol}(@var{types})} rather than just @var{symbol}. You can
15743 also use the @value{GDBN} command-line word completion facilities to list the
15744 available choices, or to finish the type list for you.
15745 @xref{Completion,, Command Completion}, for details on how to do this.
15747 @item @r{Breakpoints in functions with ABI tags}
15749 The GNU C@t{++} compiler introduced the notion of ABI ``tags'', which
15750 correspond to changes in the ABI of a type, function, or variable that
15751 would not otherwise be reflected in a mangled name. See
15752 @url{https://developers.redhat.com/blog/2015/02/05/gcc5-and-the-c11-abi/}
15755 The ABI tags are visible in C@t{++} demangled names. For example, a
15756 function that returns a std::string:
15759 std::string function(int);
15763 when compiled for the C++11 ABI is marked with the @code{cxx11} ABI
15764 tag, and @value{GDBN} displays the symbol like this:
15767 function[abi:cxx11](int)
15770 You can set a breakpoint on such functions simply as if they had no
15774 (gdb) b function(int)
15775 Breakpoint 2 at 0x40060d: file main.cc, line 10.
15776 (gdb) info breakpoints
15777 Num Type Disp Enb Address What
15778 1 breakpoint keep y 0x0040060d in function[abi:cxx11](int)
15782 On the rare occasion you need to disambiguate between different ABI
15783 tags, you can do so by simply including the ABI tag in the function
15787 (@value{GDBP}) b ambiguous[abi:other_tag](int)
15791 @node Decimal Floating Point
15792 @subsubsection Decimal Floating Point format
15793 @cindex decimal floating point format
15795 @value{GDBN} can examine, set and perform computations with numbers in
15796 decimal floating point format, which in the C language correspond to the
15797 @code{_Decimal32}, @code{_Decimal64} and @code{_Decimal128} types as
15798 specified by the extension to support decimal floating-point arithmetic.
15800 There are two encodings in use, depending on the architecture: BID (Binary
15801 Integer Decimal) for x86 and x86-64, and DPD (Densely Packed Decimal) for
15802 PowerPC and S/390. @value{GDBN} will use the appropriate encoding for the
15805 Because of a limitation in @file{libdecnumber}, the library used by @value{GDBN}
15806 to manipulate decimal floating point numbers, it is not possible to convert
15807 (using a cast, for example) integers wider than 32-bit to decimal float.
15809 In addition, in order to imitate @value{GDBN}'s behaviour with binary floating
15810 point computations, error checking in decimal float operations ignores
15811 underflow, overflow and divide by zero exceptions.
15813 In the PowerPC architecture, @value{GDBN} provides a set of pseudo-registers
15814 to inspect @code{_Decimal128} values stored in floating point registers.
15815 See @ref{PowerPC,,PowerPC} for more details.
15821 @value{GDBN} can be used to debug programs written in D and compiled with
15822 GDC, LDC or DMD compilers. Currently @value{GDBN} supports only one D
15823 specific feature --- dynamic arrays.
15828 @cindex Go (programming language)
15829 @value{GDBN} can be used to debug programs written in Go and compiled with
15830 @file{gccgo} or @file{6g} compilers.
15832 Here is a summary of the Go-specific features and restrictions:
15835 @cindex current Go package
15836 @item The current Go package
15837 The name of the current package does not need to be specified when
15838 specifying global variables and functions.
15840 For example, given the program:
15844 var myglob = "Shall we?"
15850 When stopped inside @code{main} either of these work:
15854 (gdb) p main.myglob
15857 @cindex builtin Go types
15858 @item Builtin Go types
15859 The @code{string} type is recognized by @value{GDBN} and is printed
15862 @cindex builtin Go functions
15863 @item Builtin Go functions
15864 The @value{GDBN} expression parser recognizes the @code{unsafe.Sizeof}
15865 function and handles it internally.
15867 @cindex restrictions on Go expressions
15868 @item Restrictions on Go expressions
15869 All Go operators are supported except @code{&^}.
15870 The Go @code{_} ``blank identifier'' is not supported.
15871 Automatic dereferencing of pointers is not supported.
15875 @subsection Objective-C
15877 @cindex Objective-C
15878 This section provides information about some commands and command
15879 options that are useful for debugging Objective-C code. See also
15880 @ref{Symbols, info classes}, and @ref{Symbols, info selectors}, for a
15881 few more commands specific to Objective-C support.
15884 * Method Names in Commands::
15885 * The Print Command with Objective-C::
15888 @node Method Names in Commands
15889 @subsubsection Method Names in Commands
15891 The following commands have been extended to accept Objective-C method
15892 names as line specifications:
15894 @kindex clear@r{, and Objective-C}
15895 @kindex break@r{, and Objective-C}
15896 @kindex info line@r{, and Objective-C}
15897 @kindex jump@r{, and Objective-C}
15898 @kindex list@r{, and Objective-C}
15902 @item @code{info line}
15907 A fully qualified Objective-C method name is specified as
15910 -[@var{Class} @var{methodName}]
15913 where the minus sign is used to indicate an instance method and a
15914 plus sign (not shown) is used to indicate a class method. The class
15915 name @var{Class} and method name @var{methodName} are enclosed in
15916 brackets, similar to the way messages are specified in Objective-C
15917 source code. For example, to set a breakpoint at the @code{create}
15918 instance method of class @code{Fruit} in the program currently being
15922 break -[Fruit create]
15925 To list ten program lines around the @code{initialize} class method,
15929 list +[NSText initialize]
15932 In the current version of @value{GDBN}, the plus or minus sign is
15933 required. In future versions of @value{GDBN}, the plus or minus
15934 sign will be optional, but you can use it to narrow the search. It
15935 is also possible to specify just a method name:
15941 You must specify the complete method name, including any colons. If
15942 your program's source files contain more than one @code{create} method,
15943 you'll be presented with a numbered list of classes that implement that
15944 method. Indicate your choice by number, or type @samp{0} to exit if
15947 As another example, to clear a breakpoint established at the
15948 @code{makeKeyAndOrderFront:} method of the @code{NSWindow} class, enter:
15951 clear -[NSWindow makeKeyAndOrderFront:]
15954 @node The Print Command with Objective-C
15955 @subsubsection The Print Command With Objective-C
15956 @cindex Objective-C, print objects
15957 @kindex print-object
15958 @kindex po @r{(@code{print-object})}
15960 The print command has also been extended to accept methods. For example:
15963 print -[@var{object} hash]
15966 @cindex print an Objective-C object description
15967 @cindex @code{_NSPrintForDebugger}, and printing Objective-C objects
15969 will tell @value{GDBN} to send the @code{hash} message to @var{object}
15970 and print the result. Also, an additional command has been added,
15971 @code{print-object} or @code{po} for short, which is meant to print
15972 the description of an object. However, this command may only work
15973 with certain Objective-C libraries that have a particular hook
15974 function, @code{_NSPrintForDebugger}, defined.
15977 @subsection OpenCL C
15980 This section provides information about @value{GDBN}s OpenCL C support.
15983 * OpenCL C Datatypes::
15984 * OpenCL C Expressions::
15985 * OpenCL C Operators::
15988 @node OpenCL C Datatypes
15989 @subsubsection OpenCL C Datatypes
15991 @cindex OpenCL C Datatypes
15992 @value{GDBN} supports the builtin scalar and vector datatypes specified
15993 by OpenCL 1.1. In addition the half- and double-precision floating point
15994 data types of the @code{cl_khr_fp16} and @code{cl_khr_fp64} OpenCL
15995 extensions are also known to @value{GDBN}.
15997 @node OpenCL C Expressions
15998 @subsubsection OpenCL C Expressions
16000 @cindex OpenCL C Expressions
16001 @value{GDBN} supports accesses to vector components including the access as
16002 lvalue where possible. Since OpenCL C is based on C99 most C expressions
16003 supported by @value{GDBN} can be used as well.
16005 @node OpenCL C Operators
16006 @subsubsection OpenCL C Operators
16008 @cindex OpenCL C Operators
16009 @value{GDBN} supports the operators specified by OpenCL 1.1 for scalar and
16013 @subsection Fortran
16014 @cindex Fortran-specific support in @value{GDBN}
16016 @value{GDBN} can be used to debug programs written in Fortran, but it
16017 currently supports only the features of Fortran 77 language.
16019 @cindex trailing underscore, in Fortran symbols
16020 Some Fortran compilers (@sc{gnu} Fortran 77 and Fortran 95 compilers
16021 among them) append an underscore to the names of variables and
16022 functions. When you debug programs compiled by those compilers, you
16023 will need to refer to variables and functions with a trailing
16027 * Fortran Operators:: Fortran operators and expressions
16028 * Fortran Defaults:: Default settings for Fortran
16029 * Special Fortran Commands:: Special @value{GDBN} commands for Fortran
16032 @node Fortran Operators
16033 @subsubsection Fortran Operators and Expressions
16035 @cindex Fortran operators and expressions
16037 Operators must be defined on values of specific types. For instance,
16038 @code{+} is defined on numbers, but not on characters or other non-
16039 arithmetic types. Operators are often defined on groups of types.
16043 The exponentiation operator. It raises the first operand to the power
16047 The range operator. Normally used in the form of array(low:high) to
16048 represent a section of array.
16051 The access component operator. Normally used to access elements in derived
16052 types. Also suitable for unions. As unions aren't part of regular Fortran,
16053 this can only happen when accessing a register that uses a gdbarch-defined
16057 @node Fortran Defaults
16058 @subsubsection Fortran Defaults
16060 @cindex Fortran Defaults
16062 Fortran symbols are usually case-insensitive, so @value{GDBN} by
16063 default uses case-insensitive matches for Fortran symbols. You can
16064 change that with the @samp{set case-insensitive} command, see
16065 @ref{Symbols}, for the details.
16067 @node Special Fortran Commands
16068 @subsubsection Special Fortran Commands
16070 @cindex Special Fortran commands
16072 @value{GDBN} has some commands to support Fortran-specific features,
16073 such as displaying common blocks.
16076 @cindex @code{COMMON} blocks, Fortran
16077 @kindex info common
16078 @item info common @r{[}@var{common-name}@r{]}
16079 This command prints the values contained in the Fortran @code{COMMON}
16080 block whose name is @var{common-name}. With no argument, the names of
16081 all @code{COMMON} blocks visible at the current program location are
16088 @cindex Pascal support in @value{GDBN}, limitations
16089 Debugging Pascal programs which use sets, subranges, file variables, or
16090 nested functions does not currently work. @value{GDBN} does not support
16091 entering expressions, printing values, or similar features using Pascal
16094 The Pascal-specific command @code{set print pascal_static-members}
16095 controls whether static members of Pascal objects are displayed.
16096 @xref{Print Settings, pascal_static-members}.
16101 @value{GDBN} supports the @url{https://www.rust-lang.org/, Rust
16102 Programming Language}. Type- and value-printing, and expression
16103 parsing, are reasonably complete. However, there are a few
16104 peculiarities and holes to be aware of.
16108 Linespecs (@pxref{Specify Location}) are never relative to the current
16109 crate. Instead, they act as if there were a global namespace of
16110 crates, somewhat similar to the way @code{extern crate} behaves.
16112 That is, if @value{GDBN} is stopped at a breakpoint in a function in
16113 crate @samp{A}, module @samp{B}, then @code{break B::f} will attempt
16114 to set a breakpoint in a function named @samp{f} in a crate named
16117 As a consequence of this approach, linespecs also cannot refer to
16118 items using @samp{self::} or @samp{super::}.
16121 Because @value{GDBN} implements Rust name-lookup semantics in
16122 expressions, it will sometimes prepend the current crate to a name.
16123 For example, if @value{GDBN} is stopped at a breakpoint in the crate
16124 @samp{K}, then @code{print ::x::y} will try to find the symbol
16127 However, since it is useful to be able to refer to other crates when
16128 debugging, @value{GDBN} provides the @code{extern} extension to
16129 circumvent this. To use the extension, just put @code{extern} before
16130 a path expression to refer to the otherwise unavailable ``global''
16133 In the above example, if you wanted to refer to the symbol @samp{y} in
16134 the crate @samp{x}, you would use @code{print extern x::y}.
16137 The Rust expression evaluator does not support ``statement-like''
16138 expressions such as @code{if} or @code{match}, or lambda expressions.
16141 Tuple expressions are not implemented.
16144 The Rust expression evaluator does not currently implement the
16145 @code{Drop} trait. Objects that may be created by the evaluator will
16146 never be destroyed.
16149 @value{GDBN} does not implement type inference for generics. In order
16150 to call generic functions or otherwise refer to generic items, you
16151 will have to specify the type parameters manually.
16154 @value{GDBN} currently uses the C@t{++} demangler for Rust. In most
16155 cases this does not cause any problems. However, in an expression
16156 context, completing a generic function name will give syntactically
16157 invalid results. This happens because Rust requires the @samp{::}
16158 operator between the function name and its generic arguments. For
16159 example, @value{GDBN} might provide a completion like
16160 @code{crate::f<u32>}, where the parser would require
16161 @code{crate::f::<u32>}.
16164 As of this writing, the Rust compiler (version 1.8) has a few holes in
16165 the debugging information it generates. These holes prevent certain
16166 features from being implemented by @value{GDBN}:
16170 Method calls cannot be made via traits.
16173 Operator overloading is not implemented.
16176 When debugging in a monomorphized function, you cannot use the generic
16180 The type @code{Self} is not available.
16183 @code{use} statements are not available, so some names may not be
16184 available in the crate.
16189 @subsection Modula-2
16191 @cindex Modula-2, @value{GDBN} support
16193 The extensions made to @value{GDBN} to support Modula-2 only support
16194 output from the @sc{gnu} Modula-2 compiler (which is currently being
16195 developed). Other Modula-2 compilers are not currently supported, and
16196 attempting to debug executables produced by them is most likely
16197 to give an error as @value{GDBN} reads in the executable's symbol
16200 @cindex expressions in Modula-2
16202 * M2 Operators:: Built-in operators
16203 * Built-In Func/Proc:: Built-in functions and procedures
16204 * M2 Constants:: Modula-2 constants
16205 * M2 Types:: Modula-2 types
16206 * M2 Defaults:: Default settings for Modula-2
16207 * Deviations:: Deviations from standard Modula-2
16208 * M2 Checks:: Modula-2 type and range checks
16209 * M2 Scope:: The scope operators @code{::} and @code{.}
16210 * GDB/M2:: @value{GDBN} and Modula-2
16214 @subsubsection Operators
16215 @cindex Modula-2 operators
16217 Operators must be defined on values of specific types. For instance,
16218 @code{+} is defined on numbers, but not on structures. Operators are
16219 often defined on groups of types. For the purposes of Modula-2, the
16220 following definitions hold:
16225 @emph{Integral types} consist of @code{INTEGER}, @code{CARDINAL}, and
16229 @emph{Character types} consist of @code{CHAR} and its subranges.
16232 @emph{Floating-point types} consist of @code{REAL}.
16235 @emph{Pointer types} consist of anything declared as @code{POINTER TO
16239 @emph{Scalar types} consist of all of the above.
16242 @emph{Set types} consist of @code{SET} and @code{BITSET} types.
16245 @emph{Boolean types} consist of @code{BOOLEAN}.
16249 The following operators are supported, and appear in order of
16250 increasing precedence:
16254 Function argument or array index separator.
16257 Assignment. The value of @var{var} @code{:=} @var{value} is
16261 Less than, greater than on integral, floating-point, or enumerated
16265 Less than or equal to, greater than or equal to
16266 on integral, floating-point and enumerated types, or set inclusion on
16267 set types. Same precedence as @code{<}.
16269 @item =@r{, }<>@r{, }#
16270 Equality and two ways of expressing inequality, valid on scalar types.
16271 Same precedence as @code{<}. In @value{GDBN} scripts, only @code{<>} is
16272 available for inequality, since @code{#} conflicts with the script
16276 Set membership. Defined on set types and the types of their members.
16277 Same precedence as @code{<}.
16280 Boolean disjunction. Defined on boolean types.
16283 Boolean conjunction. Defined on boolean types.
16286 The @value{GDBN} ``artificial array'' operator (@pxref{Expressions, ,Expressions}).
16289 Addition and subtraction on integral and floating-point types, or union
16290 and difference on set types.
16293 Multiplication on integral and floating-point types, or set intersection
16297 Division on floating-point types, or symmetric set difference on set
16298 types. Same precedence as @code{*}.
16301 Integer division and remainder. Defined on integral types. Same
16302 precedence as @code{*}.
16305 Negative. Defined on @code{INTEGER} and @code{REAL} data.
16308 Pointer dereferencing. Defined on pointer types.
16311 Boolean negation. Defined on boolean types. Same precedence as
16315 @code{RECORD} field selector. Defined on @code{RECORD} data. Same
16316 precedence as @code{^}.
16319 Array indexing. Defined on @code{ARRAY} data. Same precedence as @code{^}.
16322 Procedure argument list. Defined on @code{PROCEDURE} objects. Same precedence
16326 @value{GDBN} and Modula-2 scope operators.
16330 @emph{Warning:} Set expressions and their operations are not yet supported, so @value{GDBN}
16331 treats the use of the operator @code{IN}, or the use of operators
16332 @code{+}, @code{-}, @code{*}, @code{/}, @code{=}, , @code{<>}, @code{#},
16333 @code{<=}, and @code{>=} on sets as an error.
16337 @node Built-In Func/Proc
16338 @subsubsection Built-in Functions and Procedures
16339 @cindex Modula-2 built-ins
16341 Modula-2 also makes available several built-in procedures and functions.
16342 In describing these, the following metavariables are used:
16347 represents an @code{ARRAY} variable.
16350 represents a @code{CHAR} constant or variable.
16353 represents a variable or constant of integral type.
16356 represents an identifier that belongs to a set. Generally used in the
16357 same function with the metavariable @var{s}. The type of @var{s} should
16358 be @code{SET OF @var{mtype}} (where @var{mtype} is the type of @var{m}).
16361 represents a variable or constant of integral or floating-point type.
16364 represents a variable or constant of floating-point type.
16370 represents a variable.
16373 represents a variable or constant of one of many types. See the
16374 explanation of the function for details.
16377 All Modula-2 built-in procedures also return a result, described below.
16381 Returns the absolute value of @var{n}.
16384 If @var{c} is a lower case letter, it returns its upper case
16385 equivalent, otherwise it returns its argument.
16388 Returns the character whose ordinal value is @var{i}.
16391 Decrements the value in the variable @var{v} by one. Returns the new value.
16393 @item DEC(@var{v},@var{i})
16394 Decrements the value in the variable @var{v} by @var{i}. Returns the
16397 @item EXCL(@var{m},@var{s})
16398 Removes the element @var{m} from the set @var{s}. Returns the new
16401 @item FLOAT(@var{i})
16402 Returns the floating point equivalent of the integer @var{i}.
16404 @item HIGH(@var{a})
16405 Returns the index of the last member of @var{a}.
16408 Increments the value in the variable @var{v} by one. Returns the new value.
16410 @item INC(@var{v},@var{i})
16411 Increments the value in the variable @var{v} by @var{i}. Returns the
16414 @item INCL(@var{m},@var{s})
16415 Adds the element @var{m} to the set @var{s} if it is not already
16416 there. Returns the new set.
16419 Returns the maximum value of the type @var{t}.
16422 Returns the minimum value of the type @var{t}.
16425 Returns boolean TRUE if @var{i} is an odd number.
16428 Returns the ordinal value of its argument. For example, the ordinal
16429 value of a character is its @sc{ascii} value (on machines supporting
16430 the @sc{ascii} character set). The argument @var{x} must be of an
16431 ordered type, which include integral, character and enumerated types.
16433 @item SIZE(@var{x})
16434 Returns the size of its argument. The argument @var{x} can be a
16435 variable or a type.
16437 @item TRUNC(@var{r})
16438 Returns the integral part of @var{r}.
16440 @item TSIZE(@var{x})
16441 Returns the size of its argument. The argument @var{x} can be a
16442 variable or a type.
16444 @item VAL(@var{t},@var{i})
16445 Returns the member of the type @var{t} whose ordinal value is @var{i}.
16449 @emph{Warning:} Sets and their operations are not yet supported, so
16450 @value{GDBN} treats the use of procedures @code{INCL} and @code{EXCL} as
16454 @cindex Modula-2 constants
16456 @subsubsection Constants
16458 @value{GDBN} allows you to express the constants of Modula-2 in the following
16464 Integer constants are simply a sequence of digits. When used in an
16465 expression, a constant is interpreted to be type-compatible with the
16466 rest of the expression. Hexadecimal integers are specified by a
16467 trailing @samp{H}, and octal integers by a trailing @samp{B}.
16470 Floating point constants appear as a sequence of digits, followed by a
16471 decimal point and another sequence of digits. An optional exponent can
16472 then be specified, in the form @samp{E@r{[}+@r{|}-@r{]}@var{nnn}}, where
16473 @samp{@r{[}+@r{|}-@r{]}@var{nnn}} is the desired exponent. All of the
16474 digits of the floating point constant must be valid decimal (base 10)
16478 Character constants consist of a single character enclosed by a pair of
16479 like quotes, either single (@code{'}) or double (@code{"}). They may
16480 also be expressed by their ordinal value (their @sc{ascii} value, usually)
16481 followed by a @samp{C}.
16484 String constants consist of a sequence of characters enclosed by a
16485 pair of like quotes, either single (@code{'}) or double (@code{"}).
16486 Escape sequences in the style of C are also allowed. @xref{C
16487 Constants, ,C and C@t{++} Constants}, for a brief explanation of escape
16491 Enumerated constants consist of an enumerated identifier.
16494 Boolean constants consist of the identifiers @code{TRUE} and
16498 Pointer constants consist of integral values only.
16501 Set constants are not yet supported.
16505 @subsubsection Modula-2 Types
16506 @cindex Modula-2 types
16508 Currently @value{GDBN} can print the following data types in Modula-2
16509 syntax: array types, record types, set types, pointer types, procedure
16510 types, enumerated types, subrange types and base types. You can also
16511 print the contents of variables declared using these type.
16512 This section gives a number of simple source code examples together with
16513 sample @value{GDBN} sessions.
16515 The first example contains the following section of code:
16524 and you can request @value{GDBN} to interrogate the type and value of
16525 @code{r} and @code{s}.
16528 (@value{GDBP}) print s
16530 (@value{GDBP}) ptype s
16532 (@value{GDBP}) print r
16534 (@value{GDBP}) ptype r
16539 Likewise if your source code declares @code{s} as:
16543 s: SET ['A'..'Z'] ;
16547 then you may query the type of @code{s} by:
16550 (@value{GDBP}) ptype s
16551 type = SET ['A'..'Z']
16555 Note that at present you cannot interactively manipulate set
16556 expressions using the debugger.
16558 The following example shows how you might declare an array in Modula-2
16559 and how you can interact with @value{GDBN} to print its type and contents:
16563 s: ARRAY [-10..10] OF CHAR ;
16567 (@value{GDBP}) ptype s
16568 ARRAY [-10..10] OF CHAR
16571 Note that the array handling is not yet complete and although the type
16572 is printed correctly, expression handling still assumes that all
16573 arrays have a lower bound of zero and not @code{-10} as in the example
16576 Here are some more type related Modula-2 examples:
16580 colour = (blue, red, yellow, green) ;
16581 t = [blue..yellow] ;
16589 The @value{GDBN} interaction shows how you can query the data type
16590 and value of a variable.
16593 (@value{GDBP}) print s
16595 (@value{GDBP}) ptype t
16596 type = [blue..yellow]
16600 In this example a Modula-2 array is declared and its contents
16601 displayed. Observe that the contents are written in the same way as
16602 their @code{C} counterparts.
16606 s: ARRAY [1..5] OF CARDINAL ;
16612 (@value{GDBP}) print s
16613 $1 = @{1, 0, 0, 0, 0@}
16614 (@value{GDBP}) ptype s
16615 type = ARRAY [1..5] OF CARDINAL
16618 The Modula-2 language interface to @value{GDBN} also understands
16619 pointer types as shown in this example:
16623 s: POINTER TO ARRAY [1..5] OF CARDINAL ;
16630 and you can request that @value{GDBN} describes the type of @code{s}.
16633 (@value{GDBP}) ptype s
16634 type = POINTER TO ARRAY [1..5] OF CARDINAL
16637 @value{GDBN} handles compound types as we can see in this example.
16638 Here we combine array types, record types, pointer types and subrange
16649 myarray = ARRAY myrange OF CARDINAL ;
16650 myrange = [-2..2] ;
16652 s: POINTER TO ARRAY myrange OF foo ;
16656 and you can ask @value{GDBN} to describe the type of @code{s} as shown
16660 (@value{GDBP}) ptype s
16661 type = POINTER TO ARRAY [-2..2] OF foo = RECORD
16664 f3 : ARRAY [-2..2] OF CARDINAL;
16669 @subsubsection Modula-2 Defaults
16670 @cindex Modula-2 defaults
16672 If type and range checking are set automatically by @value{GDBN}, they
16673 both default to @code{on} whenever the working language changes to
16674 Modula-2. This happens regardless of whether you or @value{GDBN}
16675 selected the working language.
16677 If you allow @value{GDBN} to set the language automatically, then entering
16678 code compiled from a file whose name ends with @file{.mod} sets the
16679 working language to Modula-2. @xref{Automatically, ,Having @value{GDBN}
16680 Infer the Source Language}, for further details.
16683 @subsubsection Deviations from Standard Modula-2
16684 @cindex Modula-2, deviations from
16686 A few changes have been made to make Modula-2 programs easier to debug.
16687 This is done primarily via loosening its type strictness:
16691 Unlike in standard Modula-2, pointer constants can be formed by
16692 integers. This allows you to modify pointer variables during
16693 debugging. (In standard Modula-2, the actual address contained in a
16694 pointer variable is hidden from you; it can only be modified
16695 through direct assignment to another pointer variable or expression that
16696 returned a pointer.)
16699 C escape sequences can be used in strings and characters to represent
16700 non-printable characters. @value{GDBN} prints out strings with these
16701 escape sequences embedded. Single non-printable characters are
16702 printed using the @samp{CHR(@var{nnn})} format.
16705 The assignment operator (@code{:=}) returns the value of its right-hand
16709 All built-in procedures both modify @emph{and} return their argument.
16713 @subsubsection Modula-2 Type and Range Checks
16714 @cindex Modula-2 checks
16717 @emph{Warning:} in this release, @value{GDBN} does not yet perform type or
16720 @c FIXME remove warning when type/range checks added
16722 @value{GDBN} considers two Modula-2 variables type equivalent if:
16726 They are of types that have been declared equivalent via a @code{TYPE
16727 @var{t1} = @var{t2}} statement
16730 They have been declared on the same line. (Note: This is true of the
16731 @sc{gnu} Modula-2 compiler, but it may not be true of other compilers.)
16734 As long as type checking is enabled, any attempt to combine variables
16735 whose types are not equivalent is an error.
16737 Range checking is done on all mathematical operations, assignment, array
16738 index bounds, and all built-in functions and procedures.
16741 @subsubsection The Scope Operators @code{::} and @code{.}
16743 @cindex @code{.}, Modula-2 scope operator
16744 @cindex colon, doubled as scope operator
16746 @vindex colon-colon@r{, in Modula-2}
16747 @c Info cannot handle :: but TeX can.
16750 @vindex ::@r{, in Modula-2}
16753 There are a few subtle differences between the Modula-2 scope operator
16754 (@code{.}) and the @value{GDBN} scope operator (@code{::}). The two have
16759 @var{module} . @var{id}
16760 @var{scope} :: @var{id}
16764 where @var{scope} is the name of a module or a procedure,
16765 @var{module} the name of a module, and @var{id} is any declared
16766 identifier within your program, except another module.
16768 Using the @code{::} operator makes @value{GDBN} search the scope
16769 specified by @var{scope} for the identifier @var{id}. If it is not
16770 found in the specified scope, then @value{GDBN} searches all scopes
16771 enclosing the one specified by @var{scope}.
16773 Using the @code{.} operator makes @value{GDBN} search the current scope for
16774 the identifier specified by @var{id} that was imported from the
16775 definition module specified by @var{module}. With this operator, it is
16776 an error if the identifier @var{id} was not imported from definition
16777 module @var{module}, or if @var{id} is not an identifier in
16781 @subsubsection @value{GDBN} and Modula-2
16783 Some @value{GDBN} commands have little use when debugging Modula-2 programs.
16784 Five subcommands of @code{set print} and @code{show print} apply
16785 specifically to C and C@t{++}: @samp{vtbl}, @samp{demangle},
16786 @samp{asm-demangle}, @samp{object}, and @samp{union}. The first four
16787 apply to C@t{++}, and the last to the C @code{union} type, which has no direct
16788 analogue in Modula-2.
16790 The @code{@@} operator (@pxref{Expressions, ,Expressions}), while available
16791 with any language, is not useful with Modula-2. Its
16792 intent is to aid the debugging of @dfn{dynamic arrays}, which cannot be
16793 created in Modula-2 as they can in C or C@t{++}. However, because an
16794 address can be specified by an integral constant, the construct
16795 @samp{@{@var{type}@}@var{adrexp}} is still useful.
16797 @cindex @code{#} in Modula-2
16798 In @value{GDBN} scripts, the Modula-2 inequality operator @code{#} is
16799 interpreted as the beginning of a comment. Use @code{<>} instead.
16805 The extensions made to @value{GDBN} for Ada only support
16806 output from the @sc{gnu} Ada (GNAT) compiler.
16807 Other Ada compilers are not currently supported, and
16808 attempting to debug executables produced by them is most likely
16812 @cindex expressions in Ada
16814 * Ada Mode Intro:: General remarks on the Ada syntax
16815 and semantics supported by Ada mode
16817 * Omissions from Ada:: Restrictions on the Ada expression syntax.
16818 * Additions to Ada:: Extensions of the Ada expression syntax.
16819 * Overloading support for Ada:: Support for expressions involving overloaded
16821 * Stopping Before Main Program:: Debugging the program during elaboration.
16822 * Ada Exceptions:: Ada Exceptions
16823 * Ada Tasks:: Listing and setting breakpoints in tasks.
16824 * Ada Tasks and Core Files:: Tasking Support when Debugging Core Files
16825 * Ravenscar Profile:: Tasking Support when using the Ravenscar
16827 * Ada Settings:: New settable GDB parameters for Ada.
16828 * Ada Glitches:: Known peculiarities of Ada mode.
16831 @node Ada Mode Intro
16832 @subsubsection Introduction
16833 @cindex Ada mode, general
16835 The Ada mode of @value{GDBN} supports a fairly large subset of Ada expression
16836 syntax, with some extensions.
16837 The philosophy behind the design of this subset is
16841 That @value{GDBN} should provide basic literals and access to operations for
16842 arithmetic, dereferencing, field selection, indexing, and subprogram calls,
16843 leaving more sophisticated computations to subprograms written into the
16844 program (which therefore may be called from @value{GDBN}).
16847 That type safety and strict adherence to Ada language restrictions
16848 are not particularly important to the @value{GDBN} user.
16851 That brevity is important to the @value{GDBN} user.
16854 Thus, for brevity, the debugger acts as if all names declared in
16855 user-written packages are directly visible, even if they are not visible
16856 according to Ada rules, thus making it unnecessary to fully qualify most
16857 names with their packages, regardless of context. Where this causes
16858 ambiguity, @value{GDBN} asks the user's intent.
16860 The debugger will start in Ada mode if it detects an Ada main program.
16861 As for other languages, it will enter Ada mode when stopped in a program that
16862 was translated from an Ada source file.
16864 While in Ada mode, you may use `@t{--}' for comments. This is useful
16865 mostly for documenting command files. The standard @value{GDBN} comment
16866 (@samp{#}) still works at the beginning of a line in Ada mode, but not in the
16867 middle (to allow based literals).
16869 @node Omissions from Ada
16870 @subsubsection Omissions from Ada
16871 @cindex Ada, omissions from
16873 Here are the notable omissions from the subset:
16877 Only a subset of the attributes are supported:
16881 @t{'First}, @t{'Last}, and @t{'Length}
16882 on array objects (not on types and subtypes).
16885 @t{'Min} and @t{'Max}.
16888 @t{'Pos} and @t{'Val}.
16894 @t{'Range} on array objects (not subtypes), but only as the right
16895 operand of the membership (@code{in}) operator.
16898 @t{'Access}, @t{'Unchecked_Access}, and
16899 @t{'Unrestricted_Access} (a GNAT extension).
16907 @code{Characters.Latin_1} are not available and
16908 concatenation is not implemented. Thus, escape characters in strings are
16909 not currently available.
16912 Equality tests (@samp{=} and @samp{/=}) on arrays test for bitwise
16913 equality of representations. They will generally work correctly
16914 for strings and arrays whose elements have integer or enumeration types.
16915 They may not work correctly for arrays whose element
16916 types have user-defined equality, for arrays of real values
16917 (in particular, IEEE-conformant floating point, because of negative
16918 zeroes and NaNs), and for arrays whose elements contain unused bits with
16919 indeterminate values.
16922 The other component-by-component array operations (@code{and}, @code{or},
16923 @code{xor}, @code{not}, and relational tests other than equality)
16924 are not implemented.
16927 @cindex array aggregates (Ada)
16928 @cindex record aggregates (Ada)
16929 @cindex aggregates (Ada)
16930 There is limited support for array and record aggregates. They are
16931 permitted only on the right sides of assignments, as in these examples:
16934 (@value{GDBP}) set An_Array := (1, 2, 3, 4, 5, 6)
16935 (@value{GDBP}) set An_Array := (1, others => 0)
16936 (@value{GDBP}) set An_Array := (0|4 => 1, 1..3 => 2, 5 => 6)
16937 (@value{GDBP}) set A_2D_Array := ((1, 2, 3), (4, 5, 6), (7, 8, 9))
16938 (@value{GDBP}) set A_Record := (1, "Peter", True);
16939 (@value{GDBP}) set A_Record := (Name => "Peter", Id => 1, Alive => True)
16943 discriminant's value by assigning an aggregate has an
16944 undefined effect if that discriminant is used within the record.
16945 However, you can first modify discriminants by directly assigning to
16946 them (which normally would not be allowed in Ada), and then performing an
16947 aggregate assignment. For example, given a variable @code{A_Rec}
16948 declared to have a type such as:
16951 type Rec (Len : Small_Integer := 0) is record
16953 Vals : IntArray (1 .. Len);
16957 you can assign a value with a different size of @code{Vals} with two
16961 (@value{GDBP}) set A_Rec.Len := 4
16962 (@value{GDBP}) set A_Rec := (Id => 42, Vals => (1, 2, 3, 4))
16965 As this example also illustrates, @value{GDBN} is very loose about the usual
16966 rules concerning aggregates. You may leave out some of the
16967 components of an array or record aggregate (such as the @code{Len}
16968 component in the assignment to @code{A_Rec} above); they will retain their
16969 original values upon assignment. You may freely use dynamic values as
16970 indices in component associations. You may even use overlapping or
16971 redundant component associations, although which component values are
16972 assigned in such cases is not defined.
16975 Calls to dispatching subprograms are not implemented.
16978 The overloading algorithm is much more limited (i.e., less selective)
16979 than that of real Ada. It makes only limited use of the context in
16980 which a subexpression appears to resolve its meaning, and it is much
16981 looser in its rules for allowing type matches. As a result, some
16982 function calls will be ambiguous, and the user will be asked to choose
16983 the proper resolution.
16986 The @code{new} operator is not implemented.
16989 Entry calls are not implemented.
16992 Aside from printing, arithmetic operations on the native VAX floating-point
16993 formats are not supported.
16996 It is not possible to slice a packed array.
16999 The names @code{True} and @code{False}, when not part of a qualified name,
17000 are interpreted as if implicitly prefixed by @code{Standard}, regardless of
17002 Should your program
17003 redefine these names in a package or procedure (at best a dubious practice),
17004 you will have to use fully qualified names to access their new definitions.
17007 @node Additions to Ada
17008 @subsubsection Additions to Ada
17009 @cindex Ada, deviations from
17011 As it does for other languages, @value{GDBN} makes certain generic
17012 extensions to Ada (@pxref{Expressions}):
17016 If the expression @var{E} is a variable residing in memory (typically
17017 a local variable or array element) and @var{N} is a positive integer,
17018 then @code{@var{E}@@@var{N}} displays the values of @var{E} and the
17019 @var{N}-1 adjacent variables following it in memory as an array. In
17020 Ada, this operator is generally not necessary, since its prime use is
17021 in displaying parts of an array, and slicing will usually do this in
17022 Ada. However, there are occasional uses when debugging programs in
17023 which certain debugging information has been optimized away.
17026 @code{@var{B}::@var{var}} means ``the variable named @var{var} that
17027 appears in function or file @var{B}.'' When @var{B} is a file name,
17028 you must typically surround it in single quotes.
17031 The expression @code{@{@var{type}@} @var{addr}} means ``the variable of type
17032 @var{type} that appears at address @var{addr}.''
17035 A name starting with @samp{$} is a convenience variable
17036 (@pxref{Convenience Vars}) or a machine register (@pxref{Registers}).
17039 In addition, @value{GDBN} provides a few other shortcuts and outright
17040 additions specific to Ada:
17044 The assignment statement is allowed as an expression, returning
17045 its right-hand operand as its value. Thus, you may enter
17048 (@value{GDBP}) set x := y + 3
17049 (@value{GDBP}) print A(tmp := y + 1)
17053 The semicolon is allowed as an ``operator,'' returning as its value
17054 the value of its right-hand operand.
17055 This allows, for example,
17056 complex conditional breaks:
17059 (@value{GDBP}) break f
17060 (@value{GDBP}) condition 1 (report(i); k += 1; A(k) > 100)
17064 Rather than use catenation and symbolic character names to introduce special
17065 characters into strings, one may instead use a special bracket notation,
17066 which is also used to print strings. A sequence of characters of the form
17067 @samp{["@var{XX}"]} within a string or character literal denotes the
17068 (single) character whose numeric encoding is @var{XX} in hexadecimal. The
17069 sequence of characters @samp{["""]} also denotes a single quotation mark
17070 in strings. For example,
17072 "One line.["0a"]Next line.["0a"]"
17075 contains an ASCII newline character (@code{Ada.Characters.Latin_1.LF})
17079 The subtype used as a prefix for the attributes @t{'Pos}, @t{'Min}, and
17080 @t{'Max} is optional (and is ignored in any case). For example, it is valid
17084 (@value{GDBP}) print 'max(x, y)
17088 When printing arrays, @value{GDBN} uses positional notation when the
17089 array has a lower bound of 1, and uses a modified named notation otherwise.
17090 For example, a one-dimensional array of three integers with a lower bound
17091 of 3 might print as
17098 That is, in contrast to valid Ada, only the first component has a @code{=>}
17102 You may abbreviate attributes in expressions with any unique,
17103 multi-character subsequence of
17104 their names (an exact match gets preference).
17105 For example, you may use @t{a'len}, @t{a'gth}, or @t{a'lh}
17106 in place of @t{a'length}.
17109 @cindex quoting Ada internal identifiers
17110 Since Ada is case-insensitive, the debugger normally maps identifiers you type
17111 to lower case. The GNAT compiler uses upper-case characters for
17112 some of its internal identifiers, which are normally of no interest to users.
17113 For the rare occasions when you actually have to look at them,
17114 enclose them in angle brackets to avoid the lower-case mapping.
17117 (@value{GDBP}) print <JMPBUF_SAVE>[0]
17121 Printing an object of class-wide type or dereferencing an
17122 access-to-class-wide value will display all the components of the object's
17123 specific type (as indicated by its run-time tag). Likewise, component
17124 selection on such a value will operate on the specific type of the
17129 @node Overloading support for Ada
17130 @subsubsection Overloading support for Ada
17131 @cindex overloading, Ada
17133 The debugger supports limited overloading. Given a subprogram call in which
17134 the function symbol has multiple definitions, it will use the number of
17135 actual parameters and some information about their types to attempt to narrow
17136 the set of definitions. It also makes very limited use of context, preferring
17137 procedures to functions in the context of the @code{call} command, and
17138 functions to procedures elsewhere.
17140 If, after narrowing, the set of matching definitions still contains more than
17141 one definition, @value{GDBN} will display a menu to query which one it should
17145 (@value{GDBP}) print f(1)
17146 Multiple matches for f
17148 [1] foo.f (integer) return boolean at foo.adb:23
17149 [2] foo.f (foo.new_integer) return boolean at foo.adb:28
17153 In this case, just select one menu entry either to cancel expression evaluation
17154 (type @kbd{0} and press @key{RET}) or to continue evaluation with a specific
17155 instance (type the corresponding number and press @key{RET}).
17157 Here are a couple of commands to customize @value{GDBN}'s behavior in this
17162 @kindex set ada print-signatures
17163 @item set ada print-signatures
17164 Control whether parameter types and return types are displayed in overloads
17165 selection menus. It is @code{on} by default.
17166 @xref{Overloading support for Ada}.
17168 @kindex show ada print-signatures
17169 @item show ada print-signatures
17170 Show the current setting for displaying parameter types and return types in
17171 overloads selection menu.
17172 @xref{Overloading support for Ada}.
17176 @node Stopping Before Main Program
17177 @subsubsection Stopping at the Very Beginning
17179 @cindex breakpointing Ada elaboration code
17180 It is sometimes necessary to debug the program during elaboration, and
17181 before reaching the main procedure.
17182 As defined in the Ada Reference
17183 Manual, the elaboration code is invoked from a procedure called
17184 @code{adainit}. To run your program up to the beginning of
17185 elaboration, simply use the following two commands:
17186 @code{tbreak adainit} and @code{run}.
17188 @node Ada Exceptions
17189 @subsubsection Ada Exceptions
17191 A command is provided to list all Ada exceptions:
17194 @kindex info exceptions
17195 @item info exceptions
17196 @itemx info exceptions @var{regexp}
17197 The @code{info exceptions} command allows you to list all Ada exceptions
17198 defined within the program being debugged, as well as their addresses.
17199 With a regular expression, @var{regexp}, as argument, only those exceptions
17200 whose names match @var{regexp} are listed.
17203 Below is a small example, showing how the command can be used, first
17204 without argument, and next with a regular expression passed as an
17208 (@value{GDBP}) info exceptions
17209 All defined Ada exceptions:
17210 constraint_error: 0x613da0
17211 program_error: 0x613d20
17212 storage_error: 0x613ce0
17213 tasking_error: 0x613ca0
17214 const.aint_global_e: 0x613b00
17215 (@value{GDBP}) info exceptions const.aint
17216 All Ada exceptions matching regular expression "const.aint":
17217 constraint_error: 0x613da0
17218 const.aint_global_e: 0x613b00
17221 It is also possible to ask @value{GDBN} to stop your program's execution
17222 when an exception is raised. For more details, see @ref{Set Catchpoints}.
17225 @subsubsection Extensions for Ada Tasks
17226 @cindex Ada, tasking
17228 Support for Ada tasks is analogous to that for threads (@pxref{Threads}).
17229 @value{GDBN} provides the following task-related commands:
17234 This command shows a list of current Ada tasks, as in the following example:
17241 (@value{GDBP}) info tasks
17242 ID TID P-ID Pri State Name
17243 1 8088000 0 15 Child Activation Wait main_task
17244 2 80a4000 1 15 Accept Statement b
17245 3 809a800 1 15 Child Activation Wait a
17246 * 4 80ae800 3 15 Runnable c
17251 In this listing, the asterisk before the last task indicates it to be the
17252 task currently being inspected.
17256 Represents @value{GDBN}'s internal task number.
17262 The parent's task ID (@value{GDBN}'s internal task number).
17265 The base priority of the task.
17268 Current state of the task.
17272 The task has been created but has not been activated. It cannot be
17276 The task is not blocked for any reason known to Ada. (It may be waiting
17277 for a mutex, though.) It is conceptually "executing" in normal mode.
17280 The task is terminated, in the sense of ARM 9.3 (5). Any dependents
17281 that were waiting on terminate alternatives have been awakened and have
17282 terminated themselves.
17284 @item Child Activation Wait
17285 The task is waiting for created tasks to complete activation.
17287 @item Accept Statement
17288 The task is waiting on an accept or selective wait statement.
17290 @item Waiting on entry call
17291 The task is waiting on an entry call.
17293 @item Async Select Wait
17294 The task is waiting to start the abortable part of an asynchronous
17298 The task is waiting on a select statement with only a delay
17301 @item Child Termination Wait
17302 The task is sleeping having completed a master within itself, and is
17303 waiting for the tasks dependent on that master to become terminated or
17304 waiting on a terminate Phase.
17306 @item Wait Child in Term Alt
17307 The task is sleeping waiting for tasks on terminate alternatives to
17308 finish terminating.
17310 @item Accepting RV with @var{taskno}
17311 The task is accepting a rendez-vous with the task @var{taskno}.
17315 Name of the task in the program.
17319 @kindex info task @var{taskno}
17320 @item info task @var{taskno}
17321 This command shows detailled informations on the specified task, as in
17322 the following example:
17327 (@value{GDBP}) info tasks
17328 ID TID P-ID Pri State Name
17329 1 8077880 0 15 Child Activation Wait main_task
17330 * 2 807c468 1 15 Runnable task_1
17331 (@value{GDBP}) info task 2
17332 Ada Task: 0x807c468
17336 Parent: 1 (main_task)
17342 @kindex task@r{ (Ada)}
17343 @cindex current Ada task ID
17344 This command prints the ID of the current task.
17350 (@value{GDBP}) info tasks
17351 ID TID P-ID Pri State Name
17352 1 8077870 0 15 Child Activation Wait main_task
17353 * 2 807c458 1 15 Runnable t
17354 (@value{GDBP}) task
17355 [Current task is 2]
17358 @item task @var{taskno}
17359 @cindex Ada task switching
17360 This command is like the @code{thread @var{thread-id}}
17361 command (@pxref{Threads}). It switches the context of debugging
17362 from the current task to the given task.
17368 (@value{GDBP}) info tasks
17369 ID TID P-ID Pri State Name
17370 1 8077870 0 15 Child Activation Wait main_task
17371 * 2 807c458 1 15 Runnable t
17372 (@value{GDBP}) task 1
17373 [Switching to task 1]
17374 #0 0x8067726 in pthread_cond_wait ()
17376 #0 0x8067726 in pthread_cond_wait ()
17377 #1 0x8056714 in system.os_interface.pthread_cond_wait ()
17378 #2 0x805cb63 in system.task_primitives.operations.sleep ()
17379 #3 0x806153e in system.tasking.stages.activate_tasks ()
17380 #4 0x804aacc in un () at un.adb:5
17383 @item break @var{location} task @var{taskno}
17384 @itemx break @var{location} task @var{taskno} if @dots{}
17385 @cindex breakpoints and tasks, in Ada
17386 @cindex task breakpoints, in Ada
17387 @kindex break @dots{} task @var{taskno}@r{ (Ada)}
17388 These commands are like the @code{break @dots{} thread @dots{}}
17389 command (@pxref{Thread Stops}). The
17390 @var{location} argument specifies source lines, as described
17391 in @ref{Specify Location}.
17393 Use the qualifier @samp{task @var{taskno}} with a breakpoint command
17394 to specify that you only want @value{GDBN} to stop the program when a
17395 particular Ada task reaches this breakpoint. The @var{taskno} is one of the
17396 numeric task identifiers assigned by @value{GDBN}, shown in the first
17397 column of the @samp{info tasks} display.
17399 If you do not specify @samp{task @var{taskno}} when you set a
17400 breakpoint, the breakpoint applies to @emph{all} tasks of your
17403 You can use the @code{task} qualifier on conditional breakpoints as
17404 well; in this case, place @samp{task @var{taskno}} before the
17405 breakpoint condition (before the @code{if}).
17413 (@value{GDBP}) info tasks
17414 ID TID P-ID Pri State Name
17415 1 140022020 0 15 Child Activation Wait main_task
17416 2 140045060 1 15 Accept/Select Wait t2
17417 3 140044840 1 15 Runnable t1
17418 * 4 140056040 1 15 Runnable t3
17419 (@value{GDBP}) b 15 task 2
17420 Breakpoint 5 at 0x120044cb0: file test_task_debug.adb, line 15.
17421 (@value{GDBP}) cont
17426 Breakpoint 5, test_task_debug () at test_task_debug.adb:15
17428 (@value{GDBP}) info tasks
17429 ID TID P-ID Pri State Name
17430 1 140022020 0 15 Child Activation Wait main_task
17431 * 2 140045060 1 15 Runnable t2
17432 3 140044840 1 15 Runnable t1
17433 4 140056040 1 15 Delay Sleep t3
17437 @node Ada Tasks and Core Files
17438 @subsubsection Tasking Support when Debugging Core Files
17439 @cindex Ada tasking and core file debugging
17441 When inspecting a core file, as opposed to debugging a live program,
17442 tasking support may be limited or even unavailable, depending on
17443 the platform being used.
17444 For instance, on x86-linux, the list of tasks is available, but task
17445 switching is not supported.
17447 On certain platforms, the debugger needs to perform some
17448 memory writes in order to provide Ada tasking support. When inspecting
17449 a core file, this means that the core file must be opened with read-write
17450 privileges, using the command @samp{"set write on"} (@pxref{Patching}).
17451 Under these circumstances, you should make a backup copy of the core
17452 file before inspecting it with @value{GDBN}.
17454 @node Ravenscar Profile
17455 @subsubsection Tasking Support when using the Ravenscar Profile
17456 @cindex Ravenscar Profile
17458 The @dfn{Ravenscar Profile} is a subset of the Ada tasking features,
17459 specifically designed for systems with safety-critical real-time
17463 @kindex set ravenscar task-switching on
17464 @cindex task switching with program using Ravenscar Profile
17465 @item set ravenscar task-switching on
17466 Allows task switching when debugging a program that uses the Ravenscar
17467 Profile. This is the default.
17469 @kindex set ravenscar task-switching off
17470 @item set ravenscar task-switching off
17471 Turn off task switching when debugging a program that uses the Ravenscar
17472 Profile. This is mostly intended to disable the code that adds support
17473 for the Ravenscar Profile, in case a bug in either @value{GDBN} or in
17474 the Ravenscar runtime is preventing @value{GDBN} from working properly.
17475 To be effective, this command should be run before the program is started.
17477 @kindex show ravenscar task-switching
17478 @item show ravenscar task-switching
17479 Show whether it is possible to switch from task to task in a program
17480 using the Ravenscar Profile.
17485 @subsubsection Ada Settings
17486 @cindex Ada settings
17489 @kindex set varsize-limit
17490 @item set varsize-limit @var{size}
17491 Prevent @value{GDBN} from attempting to evaluate objects whose size
17492 is above the given limit (@var{size}) when those sizes are computed
17493 from run-time quantities. This is typically the case when the object
17494 has a variable size, such as an array whose bounds are not known at
17495 compile time for example. Setting @var{size} to @code{unlimited}
17496 removes the size limitation. By default, the limit is about 65KB.
17498 The purpose of having such a limit is to prevent @value{GDBN} from
17499 trying to grab enormous chunks of virtual memory when asked to evaluate
17500 a quantity whose bounds have been corrupted or have not yet been fully
17501 initialized. The limit applies to the results of some subexpressions
17502 as well as to complete expressions. For example, an expression denoting
17503 a simple integer component, such as @code{x.y.z}, may fail if the size of
17504 @code{x.y} is variable and exceeds @code{size}. On the other hand,
17505 @value{GDBN} is sometimes clever; the expression @code{A(i)}, where
17506 @code{A} is an array variable with non-constant size, will generally
17507 succeed regardless of the bounds on @code{A}, as long as the component
17508 size is less than @var{size}.
17510 @kindex show varsize-limit
17511 @item show varsize-limit
17512 Show the limit on types whose size is determined by run-time quantities.
17516 @subsubsection Known Peculiarities of Ada Mode
17517 @cindex Ada, problems
17519 Besides the omissions listed previously (@pxref{Omissions from Ada}),
17520 we know of several problems with and limitations of Ada mode in
17522 some of which will be fixed with planned future releases of the debugger
17523 and the GNU Ada compiler.
17527 Static constants that the compiler chooses not to materialize as objects in
17528 storage are invisible to the debugger.
17531 Named parameter associations in function argument lists are ignored (the
17532 argument lists are treated as positional).
17535 Many useful library packages are currently invisible to the debugger.
17538 Fixed-point arithmetic, conversions, input, and output is carried out using
17539 floating-point arithmetic, and may give results that only approximate those on
17543 The GNAT compiler never generates the prefix @code{Standard} for any of
17544 the standard symbols defined by the Ada language. @value{GDBN} knows about
17545 this: it will strip the prefix from names when you use it, and will never
17546 look for a name you have so qualified among local symbols, nor match against
17547 symbols in other packages or subprograms. If you have
17548 defined entities anywhere in your program other than parameters and
17549 local variables whose simple names match names in @code{Standard},
17550 GNAT's lack of qualification here can cause confusion. When this happens,
17551 you can usually resolve the confusion
17552 by qualifying the problematic names with package
17553 @code{Standard} explicitly.
17556 Older versions of the compiler sometimes generate erroneous debugging
17557 information, resulting in the debugger incorrectly printing the value
17558 of affected entities. In some cases, the debugger is able to work
17559 around an issue automatically. In other cases, the debugger is able
17560 to work around the issue, but the work-around has to be specifically
17563 @kindex set ada trust-PAD-over-XVS
17564 @kindex show ada trust-PAD-over-XVS
17567 @item set ada trust-PAD-over-XVS on
17568 Configure GDB to strictly follow the GNAT encoding when computing the
17569 value of Ada entities, particularly when @code{PAD} and @code{PAD___XVS}
17570 types are involved (see @code{ada/exp_dbug.ads} in the GCC sources for
17571 a complete description of the encoding used by the GNAT compiler).
17572 This is the default.
17574 @item set ada trust-PAD-over-XVS off
17575 This is related to the encoding using by the GNAT compiler. If @value{GDBN}
17576 sometimes prints the wrong value for certain entities, changing @code{ada
17577 trust-PAD-over-XVS} to @code{off} activates a work-around which may fix
17578 the issue. It is always safe to set @code{ada trust-PAD-over-XVS} to
17579 @code{off}, but this incurs a slight performance penalty, so it is
17580 recommended to leave this setting to @code{on} unless necessary.
17584 @cindex GNAT descriptive types
17585 @cindex GNAT encoding
17586 Internally, the debugger also relies on the compiler following a number
17587 of conventions known as the @samp{GNAT Encoding}, all documented in
17588 @file{gcc/ada/exp_dbug.ads} in the GCC sources. This encoding describes
17589 how the debugging information should be generated for certain types.
17590 In particular, this convention makes use of @dfn{descriptive types},
17591 which are artificial types generated purely to help the debugger.
17593 These encodings were defined at a time when the debugging information
17594 format used was not powerful enough to describe some of the more complex
17595 types available in Ada. Since DWARF allows us to express nearly all
17596 Ada features, the long-term goal is to slowly replace these descriptive
17597 types by their pure DWARF equivalent. To facilitate that transition,
17598 a new maintenance option is available to force the debugger to ignore
17599 those descriptive types. It allows the user to quickly evaluate how
17600 well @value{GDBN} works without them.
17604 @kindex maint ada set ignore-descriptive-types
17605 @item maintenance ada set ignore-descriptive-types [on|off]
17606 Control whether the debugger should ignore descriptive types.
17607 The default is not to ignore descriptives types (@code{off}).
17609 @kindex maint ada show ignore-descriptive-types
17610 @item maintenance ada show ignore-descriptive-types
17611 Show if descriptive types are ignored by @value{GDBN}.
17615 @node Unsupported Languages
17616 @section Unsupported Languages
17618 @cindex unsupported languages
17619 @cindex minimal language
17620 In addition to the other fully-supported programming languages,
17621 @value{GDBN} also provides a pseudo-language, called @code{minimal}.
17622 It does not represent a real programming language, but provides a set
17623 of capabilities close to what the C or assembly languages provide.
17624 This should allow most simple operations to be performed while debugging
17625 an application that uses a language currently not supported by @value{GDBN}.
17627 If the language is set to @code{auto}, @value{GDBN} will automatically
17628 select this language if the current frame corresponds to an unsupported
17632 @chapter Examining the Symbol Table
17634 The commands described in this chapter allow you to inquire about the
17635 symbols (names of variables, functions and types) defined in your
17636 program. This information is inherent in the text of your program and
17637 does not change as your program executes. @value{GDBN} finds it in your
17638 program's symbol table, in the file indicated when you started @value{GDBN}
17639 (@pxref{File Options, ,Choosing Files}), or by one of the
17640 file-management commands (@pxref{Files, ,Commands to Specify Files}).
17642 @cindex symbol names
17643 @cindex names of symbols
17644 @cindex quoting names
17645 @anchor{quoting names}
17646 Occasionally, you may need to refer to symbols that contain unusual
17647 characters, which @value{GDBN} ordinarily treats as word delimiters. The
17648 most frequent case is in referring to static variables in other
17649 source files (@pxref{Variables,,Program Variables}). File names
17650 are recorded in object files as debugging symbols, but @value{GDBN} would
17651 ordinarily parse a typical file name, like @file{foo.c}, as the three words
17652 @samp{foo} @samp{.} @samp{c}. To allow @value{GDBN} to recognize
17653 @samp{foo.c} as a single symbol, enclose it in single quotes; for example,
17660 looks up the value of @code{x} in the scope of the file @file{foo.c}.
17663 @cindex case-insensitive symbol names
17664 @cindex case sensitivity in symbol names
17665 @kindex set case-sensitive
17666 @item set case-sensitive on
17667 @itemx set case-sensitive off
17668 @itemx set case-sensitive auto
17669 Normally, when @value{GDBN} looks up symbols, it matches their names
17670 with case sensitivity determined by the current source language.
17671 Occasionally, you may wish to control that. The command @code{set
17672 case-sensitive} lets you do that by specifying @code{on} for
17673 case-sensitive matches or @code{off} for case-insensitive ones. If
17674 you specify @code{auto}, case sensitivity is reset to the default
17675 suitable for the source language. The default is case-sensitive
17676 matches for all languages except for Fortran, for which the default is
17677 case-insensitive matches.
17679 @kindex show case-sensitive
17680 @item show case-sensitive
17681 This command shows the current setting of case sensitivity for symbols
17684 @kindex set print type methods
17685 @item set print type methods
17686 @itemx set print type methods on
17687 @itemx set print type methods off
17688 Normally, when @value{GDBN} prints a class, it displays any methods
17689 declared in that class. You can control this behavior either by
17690 passing the appropriate flag to @code{ptype}, or using @command{set
17691 print type methods}. Specifying @code{on} will cause @value{GDBN} to
17692 display the methods; this is the default. Specifying @code{off} will
17693 cause @value{GDBN} to omit the methods.
17695 @kindex show print type methods
17696 @item show print type methods
17697 This command shows the current setting of method display when printing
17700 @kindex set print type nested-type-limit
17701 @item set print type nested-type-limit @var{limit}
17702 @itemx set print type nested-type-limit unlimited
17703 Set the limit of displayed nested types that the type printer will
17704 show. A @var{limit} of @code{unlimited} or @code{-1} will show all
17705 nested definitions. By default, the type printer will not show any nested
17706 types defined in classes.
17708 @kindex show print type nested-type-limit
17709 @item show print type nested-type-limit
17710 This command shows the current display limit of nested types when
17713 @kindex set print type typedefs
17714 @item set print type typedefs
17715 @itemx set print type typedefs on
17716 @itemx set print type typedefs off
17718 Normally, when @value{GDBN} prints a class, it displays any typedefs
17719 defined in that class. You can control this behavior either by
17720 passing the appropriate flag to @code{ptype}, or using @command{set
17721 print type typedefs}. Specifying @code{on} will cause @value{GDBN} to
17722 display the typedef definitions; this is the default. Specifying
17723 @code{off} will cause @value{GDBN} to omit the typedef definitions.
17724 Note that this controls whether the typedef definition itself is
17725 printed, not whether typedef names are substituted when printing other
17728 @kindex show print type typedefs
17729 @item show print type typedefs
17730 This command shows the current setting of typedef display when
17733 @kindex info address
17734 @cindex address of a symbol
17735 @item info address @var{symbol}
17736 Describe where the data for @var{symbol} is stored. For a register
17737 variable, this says which register it is kept in. For a non-register
17738 local variable, this prints the stack-frame offset at which the variable
17741 Note the contrast with @samp{print &@var{symbol}}, which does not work
17742 at all for a register variable, and for a stack local variable prints
17743 the exact address of the current instantiation of the variable.
17745 @kindex info symbol
17746 @cindex symbol from address
17747 @cindex closest symbol and offset for an address
17748 @item info symbol @var{addr}
17749 Print the name of a symbol which is stored at the address @var{addr}.
17750 If no symbol is stored exactly at @var{addr}, @value{GDBN} prints the
17751 nearest symbol and an offset from it:
17754 (@value{GDBP}) info symbol 0x54320
17755 _initialize_vx + 396 in section .text
17759 This is the opposite of the @code{info address} command. You can use
17760 it to find out the name of a variable or a function given its address.
17762 For dynamically linked executables, the name of executable or shared
17763 library containing the symbol is also printed:
17766 (@value{GDBP}) info symbol 0x400225
17767 _start + 5 in section .text of /tmp/a.out
17768 (@value{GDBP}) info symbol 0x2aaaac2811cf
17769 __read_nocancel + 6 in section .text of /usr/lib64/libc.so.6
17774 @item demangle @r{[}-l @var{language}@r{]} @r{[}@var{--}@r{]} @var{name}
17775 Demangle @var{name}.
17776 If @var{language} is provided it is the name of the language to demangle
17777 @var{name} in. Otherwise @var{name} is demangled in the current language.
17779 The @samp{--} option specifies the end of options,
17780 and is useful when @var{name} begins with a dash.
17782 The parameter @code{demangle-style} specifies how to interpret the kind
17783 of mangling used. @xref{Print Settings}.
17786 @item whatis[/@var{flags}] [@var{arg}]
17787 Print the data type of @var{arg}, which can be either an expression
17788 or a name of a data type. With no argument, print the data type of
17789 @code{$}, the last value in the value history.
17791 If @var{arg} is an expression (@pxref{Expressions, ,Expressions}), it
17792 is not actually evaluated, and any side-effecting operations (such as
17793 assignments or function calls) inside it do not take place.
17795 If @var{arg} is a variable or an expression, @code{whatis} prints its
17796 literal type as it is used in the source code. If the type was
17797 defined using a @code{typedef}, @code{whatis} will @emph{not} print
17798 the data type underlying the @code{typedef}. If the type of the
17799 variable or the expression is a compound data type, such as
17800 @code{struct} or @code{class}, @code{whatis} never prints their
17801 fields or methods. It just prints the @code{struct}/@code{class}
17802 name (a.k.a.@: its @dfn{tag}). If you want to see the members of
17803 such a compound data type, use @code{ptype}.
17805 If @var{arg} is a type name that was defined using @code{typedef},
17806 @code{whatis} @dfn{unrolls} only one level of that @code{typedef}.
17807 Unrolling means that @code{whatis} will show the underlying type used
17808 in the @code{typedef} declaration of @var{arg}. However, if that
17809 underlying type is also a @code{typedef}, @code{whatis} will not
17812 For C code, the type names may also have the form @samp{class
17813 @var{class-name}}, @samp{struct @var{struct-tag}}, @samp{union
17814 @var{union-tag}} or @samp{enum @var{enum-tag}}.
17816 @var{flags} can be used to modify how the type is displayed.
17817 Available flags are:
17821 Display in ``raw'' form. Normally, @value{GDBN} substitutes template
17822 parameters and typedefs defined in a class when printing the class'
17823 members. The @code{/r} flag disables this.
17826 Do not print methods defined in the class.
17829 Print methods defined in the class. This is the default, but the flag
17830 exists in case you change the default with @command{set print type methods}.
17833 Do not print typedefs defined in the class. Note that this controls
17834 whether the typedef definition itself is printed, not whether typedef
17835 names are substituted when printing other types.
17838 Print typedefs defined in the class. This is the default, but the flag
17839 exists in case you change the default with @command{set print type typedefs}.
17842 Print the offsets and sizes of fields in a struct, similar to what the
17843 @command{pahole} tool does. This option implies the @code{/tm} flags.
17845 For example, given the following declarations:
17881 Issuing a @kbd{ptype /o struct tuv} command would print:
17884 (@value{GDBP}) ptype /o struct tuv
17885 /* offset | size */ type = struct tuv @{
17886 /* 0 | 4 */ int a1;
17887 /* XXX 4-byte hole */
17888 /* 8 | 8 */ char *a2;
17889 /* 16 | 4 */ int a3;
17891 /* total size (bytes): 24 */
17895 Notice the format of the first column of comments. There, you can
17896 find two parts separated by the @samp{|} character: the @emph{offset},
17897 which indicates where the field is located inside the struct, in
17898 bytes, and the @emph{size} of the field. Another interesting line is
17899 the marker of a @emph{hole} in the struct, indicating that it may be
17900 possible to pack the struct and make it use less space by reorganizing
17903 It is also possible to print offsets inside an union:
17906 (@value{GDBP}) ptype /o union qwe
17907 /* offset | size */ type = union qwe @{
17908 /* 24 */ struct tuv @{
17909 /* 0 | 4 */ int a1;
17910 /* XXX 4-byte hole */
17911 /* 8 | 8 */ char *a2;
17912 /* 16 | 4 */ int a3;
17914 /* total size (bytes): 24 */
17916 /* 40 */ struct xyz @{
17917 /* 0 | 4 */ int f1;
17918 /* 4 | 1 */ char f2;
17919 /* XXX 3-byte hole */
17920 /* 8 | 8 */ void *f3;
17921 /* 16 | 24 */ struct tuv @{
17922 /* 16 | 4 */ int a1;
17923 /* XXX 4-byte hole */
17924 /* 24 | 8 */ char *a2;
17925 /* 32 | 4 */ int a3;
17927 /* total size (bytes): 24 */
17930 /* total size (bytes): 40 */
17933 /* total size (bytes): 40 */
17937 In this case, since @code{struct tuv} and @code{struct xyz} occupy the
17938 same space (because we are dealing with an union), the offset is not
17939 printed for them. However, you can still examine the offset of each
17940 of these structures' fields.
17942 Another useful scenario is printing the offsets of a struct containing
17946 (@value{GDBP}) ptype /o struct tyu
17947 /* offset | size */ type = struct tyu @{
17948 /* 0:31 | 4 */ int a1 : 1;
17949 /* 0:28 | 4 */ int a2 : 3;
17950 /* 0: 5 | 4 */ int a3 : 23;
17951 /* 3: 3 | 1 */ signed char a4 : 2;
17952 /* XXX 3-bit hole */
17953 /* XXX 4-byte hole */
17954 /* 8 | 8 */ int64_t a5;
17955 /* 16: 0 | 4 */ int a6 : 5;
17956 /* 16: 5 | 8 */ int64_t a7 : 3;
17957 "/* XXX 7-byte padding */
17959 /* total size (bytes): 24 */
17963 Note how the offset information is now extended to also include the
17964 first bit of the bitfield.
17968 @item ptype[/@var{flags}] [@var{arg}]
17969 @code{ptype} accepts the same arguments as @code{whatis}, but prints a
17970 detailed description of the type, instead of just the name of the type.
17971 @xref{Expressions, ,Expressions}.
17973 Contrary to @code{whatis}, @code{ptype} always unrolls any
17974 @code{typedef}s in its argument declaration, whether the argument is
17975 a variable, expression, or a data type. This means that @code{ptype}
17976 of a variable or an expression will not print literally its type as
17977 present in the source code---use @code{whatis} for that. @code{typedef}s at
17978 the pointer or reference targets are also unrolled. Only @code{typedef}s of
17979 fields, methods and inner @code{class typedef}s of @code{struct}s,
17980 @code{class}es and @code{union}s are not unrolled even with @code{ptype}.
17982 For example, for this variable declaration:
17985 typedef double real_t;
17986 struct complex @{ real_t real; double imag; @};
17987 typedef struct complex complex_t;
17989 real_t *real_pointer_var;
17993 the two commands give this output:
17997 (@value{GDBP}) whatis var
17999 (@value{GDBP}) ptype var
18000 type = struct complex @{
18004 (@value{GDBP}) whatis complex_t
18005 type = struct complex
18006 (@value{GDBP}) whatis struct complex
18007 type = struct complex
18008 (@value{GDBP}) ptype struct complex
18009 type = struct complex @{
18013 (@value{GDBP}) whatis real_pointer_var
18015 (@value{GDBP}) ptype real_pointer_var
18021 As with @code{whatis}, using @code{ptype} without an argument refers to
18022 the type of @code{$}, the last value in the value history.
18024 @cindex incomplete type
18025 Sometimes, programs use opaque data types or incomplete specifications
18026 of complex data structure. If the debug information included in the
18027 program does not allow @value{GDBN} to display a full declaration of
18028 the data type, it will say @samp{<incomplete type>}. For example,
18029 given these declarations:
18033 struct foo *fooptr;
18037 but no definition for @code{struct foo} itself, @value{GDBN} will say:
18040 (@value{GDBP}) ptype foo
18041 $1 = <incomplete type>
18045 ``Incomplete type'' is C terminology for data types that are not
18046 completely specified.
18048 @cindex unknown type
18049 Othertimes, information about a variable's type is completely absent
18050 from the debug information included in the program. This most often
18051 happens when the program or library where the variable is defined
18052 includes no debug information at all. @value{GDBN} knows the variable
18053 exists from inspecting the linker/loader symbol table (e.g., the ELF
18054 dynamic symbol table), but such symbols do not contain type
18055 information. Inspecting the type of a (global) variable for which
18056 @value{GDBN} has no type information shows:
18059 (@value{GDBP}) ptype var
18060 type = <data variable, no debug info>
18063 @xref{Variables, no debug info variables}, for how to print the values
18067 @item info types @var{regexp}
18069 Print a brief description of all types whose names match the regular
18070 expression @var{regexp} (or all types in your program, if you supply
18071 no argument). Each complete typename is matched as though it were a
18072 complete line; thus, @samp{i type value} gives information on all
18073 types in your program whose names include the string @code{value}, but
18074 @samp{i type ^value$} gives information only on types whose complete
18075 name is @code{value}.
18077 In programs using different languages, @value{GDBN} chooses the syntax
18078 to print the type description according to the
18079 @samp{set language} value: using @samp{set language auto}
18080 (see @ref{Automatically, ,Set Language Automatically}) means to use the
18081 language of the type, other values mean to use
18082 the manually specified language (see @ref{Manually, ,Set Language Manually}).
18084 This command differs from @code{ptype} in two ways: first, like
18085 @code{whatis}, it does not print a detailed description; second, it
18086 lists all source files and line numbers where a type is defined.
18088 @kindex info type-printers
18089 @item info type-printers
18090 Versions of @value{GDBN} that ship with Python scripting enabled may
18091 have ``type printers'' available. When using @command{ptype} or
18092 @command{whatis}, these printers are consulted when the name of a type
18093 is needed. @xref{Type Printing API}, for more information on writing
18096 @code{info type-printers} displays all the available type printers.
18098 @kindex enable type-printer
18099 @kindex disable type-printer
18100 @item enable type-printer @var{name}@dots{}
18101 @item disable type-printer @var{name}@dots{}
18102 These commands can be used to enable or disable type printers.
18105 @cindex local variables
18106 @item info scope @var{location}
18107 List all the variables local to a particular scope. This command
18108 accepts a @var{location} argument---a function name, a source line, or
18109 an address preceded by a @samp{*}, and prints all the variables local
18110 to the scope defined by that location. (@xref{Specify Location}, for
18111 details about supported forms of @var{location}.) For example:
18114 (@value{GDBP}) @b{info scope command_line_handler}
18115 Scope for command_line_handler:
18116 Symbol rl is an argument at stack/frame offset 8, length 4.
18117 Symbol linebuffer is in static storage at address 0x150a18, length 4.
18118 Symbol linelength is in static storage at address 0x150a1c, length 4.
18119 Symbol p is a local variable in register $esi, length 4.
18120 Symbol p1 is a local variable in register $ebx, length 4.
18121 Symbol nline is a local variable in register $edx, length 4.
18122 Symbol repeat is a local variable at frame offset -8, length 4.
18126 This command is especially useful for determining what data to collect
18127 during a @dfn{trace experiment}, see @ref{Tracepoint Actions,
18130 @kindex info source
18132 Show information about the current source file---that is, the source file for
18133 the function containing the current point of execution:
18136 the name of the source file, and the directory containing it,
18138 the directory it was compiled in,
18140 its length, in lines,
18142 which programming language it is written in,
18144 if the debug information provides it, the program that compiled the file
18145 (which may include, e.g., the compiler version and command line arguments),
18147 whether the executable includes debugging information for that file, and
18148 if so, what format the information is in (e.g., STABS, Dwarf 2, etc.), and
18150 whether the debugging information includes information about
18151 preprocessor macros.
18155 @kindex info sources
18157 Print the names of all source files in your program for which there is
18158 debugging information, organized into two lists: files whose symbols
18159 have already been read, and files whose symbols will be read when needed.
18161 @kindex info functions
18162 @item info functions [-q]
18163 Print the names and data types of all defined functions.
18164 Similarly to @samp{info types}, this command groups its output by source
18165 files and annotates each function definition with its source line
18168 In programs using different languages, @value{GDBN} chooses the syntax
18169 to print the function name and type according to the
18170 @samp{set language} value: using @samp{set language auto}
18171 (see @ref{Automatically, ,Set Language Automatically}) means to use the
18172 language of the function, other values mean to use
18173 the manually specified language (see @ref{Manually, ,Set Language Manually}).
18175 The optional flag @samp{-q}, which stands for @samp{quiet}, disables
18176 printing header information and messages explaining why no functions
18179 @item info functions [-q] [-t @var{type_regexp}] [@var{regexp}]
18180 Like @samp{info functions}, but only print the names and data types
18181 of the functions selected with the provided regexp(s).
18183 If @var{regexp} is provided, print only the functions whose names
18184 match the regular expression @var{regexp}.
18185 Thus, @samp{info fun step} finds all functions whose
18186 names include @code{step}; @samp{info fun ^step} finds those whose names
18187 start with @code{step}. If a function name contains characters that
18188 conflict with the regular expression language (e.g.@:
18189 @samp{operator*()}), they may be quoted with a backslash.
18191 If @var{type_regexp} is provided, print only the functions whose
18192 types, as printed by the @code{whatis} command, match
18193 the regular expression @var{type_regexp}.
18194 If @var{type_regexp} contains space(s), it should be enclosed in
18195 quote characters. If needed, use backslash to escape the meaning
18196 of special characters or quotes.
18197 Thus, @samp{info fun -t '^int ('} finds the functions that return
18198 an integer; @samp{info fun -t '(.*int.*'} finds the functions that
18199 have an argument type containing int; @samp{info fun -t '^int (' ^step}
18200 finds the functions whose names start with @code{step} and that return
18203 If both @var{regexp} and @var{type_regexp} are provided, a function
18204 is printed only if its name matches @var{regexp} and its type matches
18208 @kindex info variables
18209 @item info variables [-q]
18210 Print the names and data types of all variables that are defined
18211 outside of functions (i.e.@: excluding local variables).
18212 The printed variables are grouped by source files and annotated with
18213 their respective source line numbers.
18215 In programs using different languages, @value{GDBN} chooses the syntax
18216 to print the variable name and type according to the
18217 @samp{set language} value: using @samp{set language auto}
18218 (see @ref{Automatically, ,Set Language Automatically}) means to use the
18219 language of the variable, other values mean to use
18220 the manually specified language (see @ref{Manually, ,Set Language Manually}).
18222 The optional flag @samp{-q}, which stands for @samp{quiet}, disables
18223 printing header information and messages explaining why no variables
18226 @item info variables [-q] [-t @var{type_regexp}] [@var{regexp}]
18227 Like @kbd{info variables}, but only print the variables selected
18228 with the provided regexp(s).
18230 If @var{regexp} is provided, print only the variables whose names
18231 match the regular expression @var{regexp}.
18233 If @var{type_regexp} is provided, print only the variables whose
18234 types, as printed by the @code{whatis} command, match
18235 the regular expression @var{type_regexp}.
18236 If @var{type_regexp} contains space(s), it should be enclosed in
18237 quote characters. If needed, use backslash to escape the meaning
18238 of special characters or quotes.
18240 If both @var{regexp} and @var{type_regexp} are provided, an argument
18241 is printed only if its name matches @var{regexp} and its type matches
18244 @kindex info classes
18245 @cindex Objective-C, classes and selectors
18247 @itemx info classes @var{regexp}
18248 Display all Objective-C classes in your program, or
18249 (with the @var{regexp} argument) all those matching a particular regular
18252 @kindex info selectors
18253 @item info selectors
18254 @itemx info selectors @var{regexp}
18255 Display all Objective-C selectors in your program, or
18256 (with the @var{regexp} argument) all those matching a particular regular
18260 This was never implemented.
18261 @kindex info methods
18263 @itemx info methods @var{regexp}
18264 The @code{info methods} command permits the user to examine all defined
18265 methods within C@t{++} program, or (with the @var{regexp} argument) a
18266 specific set of methods found in the various C@t{++} classes. Many
18267 C@t{++} classes provide a large number of methods. Thus, the output
18268 from the @code{ptype} command can be overwhelming and hard to use. The
18269 @code{info-methods} command filters the methods, printing only those
18270 which match the regular-expression @var{regexp}.
18273 @cindex opaque data types
18274 @kindex set opaque-type-resolution
18275 @item set opaque-type-resolution on
18276 Tell @value{GDBN} to resolve opaque types. An opaque type is a type
18277 declared as a pointer to a @code{struct}, @code{class}, or
18278 @code{union}---for example, @code{struct MyType *}---that is used in one
18279 source file although the full declaration of @code{struct MyType} is in
18280 another source file. The default is on.
18282 A change in the setting of this subcommand will not take effect until
18283 the next time symbols for a file are loaded.
18285 @item set opaque-type-resolution off
18286 Tell @value{GDBN} not to resolve opaque types. In this case, the type
18287 is printed as follows:
18289 @{<no data fields>@}
18292 @kindex show opaque-type-resolution
18293 @item show opaque-type-resolution
18294 Show whether opaque types are resolved or not.
18296 @kindex set print symbol-loading
18297 @cindex print messages when symbols are loaded
18298 @item set print symbol-loading
18299 @itemx set print symbol-loading full
18300 @itemx set print symbol-loading brief
18301 @itemx set print symbol-loading off
18302 The @code{set print symbol-loading} command allows you to control the
18303 printing of messages when @value{GDBN} loads symbol information.
18304 By default a message is printed for the executable and one for each
18305 shared library, and normally this is what you want. However, when
18306 debugging apps with large numbers of shared libraries these messages
18308 When set to @code{brief} a message is printed for each executable,
18309 and when @value{GDBN} loads a collection of shared libraries at once
18310 it will only print one message regardless of the number of shared
18311 libraries. When set to @code{off} no messages are printed.
18313 @kindex show print symbol-loading
18314 @item show print symbol-loading
18315 Show whether messages will be printed when a @value{GDBN} command
18316 entered from the keyboard causes symbol information to be loaded.
18318 @kindex maint print symbols
18319 @cindex symbol dump
18320 @kindex maint print psymbols
18321 @cindex partial symbol dump
18322 @kindex maint print msymbols
18323 @cindex minimal symbol dump
18324 @item maint print symbols @r{[}-pc @var{address}@r{]} @r{[}@var{filename}@r{]}
18325 @itemx maint print symbols @r{[}-objfile @var{objfile}@r{]} @r{[}-source @var{source}@r{]} @r{[}--@r{]} @r{[}@var{filename}@r{]}
18326 @itemx maint print psymbols @r{[}-objfile @var{objfile}@r{]} @r{[}-pc @var{address}@r{]} @r{[}--@r{]} @r{[}@var{filename}@r{]}
18327 @itemx maint print psymbols @r{[}-objfile @var{objfile}@r{]} @r{[}-source @var{source}@r{]} @r{[}--@r{]} @r{[}@var{filename}@r{]}
18328 @itemx maint print msymbols @r{[}-objfile @var{objfile}@r{]} @r{[}--@r{]} @r{[}@var{filename}@r{]}
18329 Write a dump of debugging symbol data into the file @var{filename} or
18330 the terminal if @var{filename} is unspecified.
18331 If @code{-objfile @var{objfile}} is specified, only dump symbols for
18333 If @code{-pc @var{address}} is specified, only dump symbols for the file
18334 with code at that address. Note that @var{address} may be a symbol like
18336 If @code{-source @var{source}} is specified, only dump symbols for that
18339 These commands are used to debug the @value{GDBN} symbol-reading code.
18340 These commands do not modify internal @value{GDBN} state, therefore
18341 @samp{maint print symbols} will only print symbols for already expanded symbol
18343 You can use the command @code{info sources} to find out which files these are.
18344 If you use @samp{maint print psymbols} instead, the dump shows information
18345 about symbols that @value{GDBN} only knows partially---that is, symbols
18346 defined in files that @value{GDBN} has skimmed, but not yet read completely.
18347 Finally, @samp{maint print msymbols} just dumps ``minimal symbols'', e.g.,
18350 @xref{Files, ,Commands to Specify Files}, for a discussion of how
18351 @value{GDBN} reads symbols (in the description of @code{symbol-file}).
18353 @kindex maint info symtabs
18354 @kindex maint info psymtabs
18355 @cindex listing @value{GDBN}'s internal symbol tables
18356 @cindex symbol tables, listing @value{GDBN}'s internal
18357 @cindex full symbol tables, listing @value{GDBN}'s internal
18358 @cindex partial symbol tables, listing @value{GDBN}'s internal
18359 @item maint info symtabs @r{[} @var{regexp} @r{]}
18360 @itemx maint info psymtabs @r{[} @var{regexp} @r{]}
18362 List the @code{struct symtab} or @code{struct partial_symtab}
18363 structures whose names match @var{regexp}. If @var{regexp} is not
18364 given, list them all. The output includes expressions which you can
18365 copy into a @value{GDBN} debugging this one to examine a particular
18366 structure in more detail. For example:
18369 (@value{GDBP}) maint info psymtabs dwarf2read
18370 @{ objfile /home/gnu/build/gdb/gdb
18371 ((struct objfile *) 0x82e69d0)
18372 @{ psymtab /home/gnu/src/gdb/dwarf2read.c
18373 ((struct partial_symtab *) 0x8474b10)
18376 text addresses 0x814d3c8 -- 0x8158074
18377 globals (* (struct partial_symbol **) 0x8507a08 @@ 9)
18378 statics (* (struct partial_symbol **) 0x40e95b78 @@ 2882)
18379 dependencies (none)
18382 (@value{GDBP}) maint info symtabs
18386 We see that there is one partial symbol table whose filename contains
18387 the string @samp{dwarf2read}, belonging to the @samp{gdb} executable;
18388 and we see that @value{GDBN} has not read in any symtabs yet at all.
18389 If we set a breakpoint on a function, that will cause @value{GDBN} to
18390 read the symtab for the compilation unit containing that function:
18393 (@value{GDBP}) break dwarf2_psymtab_to_symtab
18394 Breakpoint 1 at 0x814e5da: file /home/gnu/src/gdb/dwarf2read.c,
18396 (@value{GDBP}) maint info symtabs
18397 @{ objfile /home/gnu/build/gdb/gdb
18398 ((struct objfile *) 0x82e69d0)
18399 @{ symtab /home/gnu/src/gdb/dwarf2read.c
18400 ((struct symtab *) 0x86c1f38)
18403 blockvector ((struct blockvector *) 0x86c1bd0) (primary)
18404 linetable ((struct linetable *) 0x8370fa0)
18405 debugformat DWARF 2
18411 @kindex maint info line-table
18412 @cindex listing @value{GDBN}'s internal line tables
18413 @cindex line tables, listing @value{GDBN}'s internal
18414 @item maint info line-table @r{[} @var{regexp} @r{]}
18416 List the @code{struct linetable} from all @code{struct symtab}
18417 instances whose name matches @var{regexp}. If @var{regexp} is not
18418 given, list the @code{struct linetable} from all @code{struct symtab}.
18420 @kindex maint set symbol-cache-size
18421 @cindex symbol cache size
18422 @item maint set symbol-cache-size @var{size}
18423 Set the size of the symbol cache to @var{size}.
18424 The default size is intended to be good enough for debugging
18425 most applications. This option exists to allow for experimenting
18426 with different sizes.
18428 @kindex maint show symbol-cache-size
18429 @item maint show symbol-cache-size
18430 Show the size of the symbol cache.
18432 @kindex maint print symbol-cache
18433 @cindex symbol cache, printing its contents
18434 @item maint print symbol-cache
18435 Print the contents of the symbol cache.
18436 This is useful when debugging symbol cache issues.
18438 @kindex maint print symbol-cache-statistics
18439 @cindex symbol cache, printing usage statistics
18440 @item maint print symbol-cache-statistics
18441 Print symbol cache usage statistics.
18442 This helps determine how well the cache is being utilized.
18444 @kindex maint flush-symbol-cache
18445 @cindex symbol cache, flushing
18446 @item maint flush-symbol-cache
18447 Flush the contents of the symbol cache, all entries are removed.
18448 This command is useful when debugging the symbol cache.
18449 It is also useful when collecting performance data.
18454 @chapter Altering Execution
18456 Once you think you have found an error in your program, you might want to
18457 find out for certain whether correcting the apparent error would lead to
18458 correct results in the rest of the run. You can find the answer by
18459 experiment, using the @value{GDBN} features for altering execution of the
18462 For example, you can store new values into variables or memory
18463 locations, give your program a signal, restart it at a different
18464 address, or even return prematurely from a function.
18467 * Assignment:: Assignment to variables
18468 * Jumping:: Continuing at a different address
18469 * Signaling:: Giving your program a signal
18470 * Returning:: Returning from a function
18471 * Calling:: Calling your program's functions
18472 * Patching:: Patching your program
18473 * Compiling and Injecting Code:: Compiling and injecting code in @value{GDBN}
18477 @section Assignment to Variables
18480 @cindex setting variables
18481 To alter the value of a variable, evaluate an assignment expression.
18482 @xref{Expressions, ,Expressions}. For example,
18489 stores the value 4 into the variable @code{x}, and then prints the
18490 value of the assignment expression (which is 4).
18491 @xref{Languages, ,Using @value{GDBN} with Different Languages}, for more
18492 information on operators in supported languages.
18494 @kindex set variable
18495 @cindex variables, setting
18496 If you are not interested in seeing the value of the assignment, use the
18497 @code{set} command instead of the @code{print} command. @code{set} is
18498 really the same as @code{print} except that the expression's value is
18499 not printed and is not put in the value history (@pxref{Value History,
18500 ,Value History}). The expression is evaluated only for its effects.
18502 If the beginning of the argument string of the @code{set} command
18503 appears identical to a @code{set} subcommand, use the @code{set
18504 variable} command instead of just @code{set}. This command is identical
18505 to @code{set} except for its lack of subcommands. For example, if your
18506 program has a variable @code{width}, you get an error if you try to set
18507 a new value with just @samp{set width=13}, because @value{GDBN} has the
18508 command @code{set width}:
18511 (@value{GDBP}) whatis width
18513 (@value{GDBP}) p width
18515 (@value{GDBP}) set width=47
18516 Invalid syntax in expression.
18520 The invalid expression, of course, is @samp{=47}. In
18521 order to actually set the program's variable @code{width}, use
18524 (@value{GDBP}) set var width=47
18527 Because the @code{set} command has many subcommands that can conflict
18528 with the names of program variables, it is a good idea to use the
18529 @code{set variable} command instead of just @code{set}. For example, if
18530 your program has a variable @code{g}, you run into problems if you try
18531 to set a new value with just @samp{set g=4}, because @value{GDBN} has
18532 the command @code{set gnutarget}, abbreviated @code{set g}:
18536 (@value{GDBP}) whatis g
18540 (@value{GDBP}) set g=4
18544 The program being debugged has been started already.
18545 Start it from the beginning? (y or n) y
18546 Starting program: /home/smith/cc_progs/a.out
18547 "/home/smith/cc_progs/a.out": can't open to read symbols:
18548 Invalid bfd target.
18549 (@value{GDBP}) show g
18550 The current BFD target is "=4".
18555 The program variable @code{g} did not change, and you silently set the
18556 @code{gnutarget} to an invalid value. In order to set the variable
18560 (@value{GDBP}) set var g=4
18563 @value{GDBN} allows more implicit conversions in assignments than C; you can
18564 freely store an integer value into a pointer variable or vice versa,
18565 and you can convert any structure to any other structure that is the
18566 same length or shorter.
18567 @comment FIXME: how do structs align/pad in these conversions?
18568 @comment /doc@cygnus.com 18dec1990
18570 To store values into arbitrary places in memory, use the @samp{@{@dots{}@}}
18571 construct to generate a value of specified type at a specified address
18572 (@pxref{Expressions, ,Expressions}). For example, @code{@{int@}0x83040} refers
18573 to memory location @code{0x83040} as an integer (which implies a certain size
18574 and representation in memory), and
18577 set @{int@}0x83040 = 4
18581 stores the value 4 into that memory location.
18584 @section Continuing at a Different Address
18586 Ordinarily, when you continue your program, you do so at the place where
18587 it stopped, with the @code{continue} command. You can instead continue at
18588 an address of your own choosing, with the following commands:
18592 @kindex j @r{(@code{jump})}
18593 @item jump @var{location}
18594 @itemx j @var{location}
18595 Resume execution at @var{location}. Execution stops again immediately
18596 if there is a breakpoint there. @xref{Specify Location}, for a description
18597 of the different forms of @var{location}. It is common
18598 practice to use the @code{tbreak} command in conjunction with
18599 @code{jump}. @xref{Set Breaks, ,Setting Breakpoints}.
18601 The @code{jump} command does not change the current stack frame, or
18602 the stack pointer, or the contents of any memory location or any
18603 register other than the program counter. If @var{location} is in
18604 a different function from the one currently executing, the results may
18605 be bizarre if the two functions expect different patterns of arguments or
18606 of local variables. For this reason, the @code{jump} command requests
18607 confirmation if the specified line is not in the function currently
18608 executing. However, even bizarre results are predictable if you are
18609 well acquainted with the machine-language code of your program.
18612 On many systems, you can get much the same effect as the @code{jump}
18613 command by storing a new value into the register @code{$pc}. The
18614 difference is that this does not start your program running; it only
18615 changes the address of where it @emph{will} run when you continue. For
18623 makes the next @code{continue} command or stepping command execute at
18624 address @code{0x485}, rather than at the address where your program stopped.
18625 @xref{Continuing and Stepping, ,Continuing and Stepping}.
18627 The most common occasion to use the @code{jump} command is to back
18628 up---perhaps with more breakpoints set---over a portion of a program
18629 that has already executed, in order to examine its execution in more
18634 @section Giving your Program a Signal
18635 @cindex deliver a signal to a program
18639 @item signal @var{signal}
18640 Resume execution where your program is stopped, but immediately give it the
18641 signal @var{signal}. The @var{signal} can be the name or the number of a
18642 signal. For example, on many systems @code{signal 2} and @code{signal
18643 SIGINT} are both ways of sending an interrupt signal.
18645 Alternatively, if @var{signal} is zero, continue execution without
18646 giving a signal. This is useful when your program stopped on account of
18647 a signal and would ordinarily see the signal when resumed with the
18648 @code{continue} command; @samp{signal 0} causes it to resume without a
18651 @emph{Note:} When resuming a multi-threaded program, @var{signal} is
18652 delivered to the currently selected thread, not the thread that last
18653 reported a stop. This includes the situation where a thread was
18654 stopped due to a signal. So if you want to continue execution
18655 suppressing the signal that stopped a thread, you should select that
18656 same thread before issuing the @samp{signal 0} command. If you issue
18657 the @samp{signal 0} command with another thread as the selected one,
18658 @value{GDBN} detects that and asks for confirmation.
18660 Invoking the @code{signal} command is not the same as invoking the
18661 @code{kill} utility from the shell. Sending a signal with @code{kill}
18662 causes @value{GDBN} to decide what to do with the signal depending on
18663 the signal handling tables (@pxref{Signals}). The @code{signal} command
18664 passes the signal directly to your program.
18666 @code{signal} does not repeat when you press @key{RET} a second time
18667 after executing the command.
18669 @kindex queue-signal
18670 @item queue-signal @var{signal}
18671 Queue @var{signal} to be delivered immediately to the current thread
18672 when execution of the thread resumes. The @var{signal} can be the name or
18673 the number of a signal. For example, on many systems @code{signal 2} and
18674 @code{signal SIGINT} are both ways of sending an interrupt signal.
18675 The handling of the signal must be set to pass the signal to the program,
18676 otherwise @value{GDBN} will report an error.
18677 You can control the handling of signals from @value{GDBN} with the
18678 @code{handle} command (@pxref{Signals}).
18680 Alternatively, if @var{signal} is zero, any currently queued signal
18681 for the current thread is discarded and when execution resumes no signal
18682 will be delivered. This is useful when your program stopped on account
18683 of a signal and would ordinarily see the signal when resumed with the
18684 @code{continue} command.
18686 This command differs from the @code{signal} command in that the signal
18687 is just queued, execution is not resumed. And @code{queue-signal} cannot
18688 be used to pass a signal whose handling state has been set to @code{nopass}
18693 @xref{stepping into signal handlers}, for information on how stepping
18694 commands behave when the thread has a signal queued.
18697 @section Returning from a Function
18700 @cindex returning from a function
18703 @itemx return @var{expression}
18704 You can cancel execution of a function call with the @code{return}
18705 command. If you give an
18706 @var{expression} argument, its value is used as the function's return
18710 When you use @code{return}, @value{GDBN} discards the selected stack frame
18711 (and all frames within it). You can think of this as making the
18712 discarded frame return prematurely. If you wish to specify a value to
18713 be returned, give that value as the argument to @code{return}.
18715 This pops the selected stack frame (@pxref{Selection, ,Selecting a
18716 Frame}), and any other frames inside of it, leaving its caller as the
18717 innermost remaining frame. That frame becomes selected. The
18718 specified value is stored in the registers used for returning values
18721 The @code{return} command does not resume execution; it leaves the
18722 program stopped in the state that would exist if the function had just
18723 returned. In contrast, the @code{finish} command (@pxref{Continuing
18724 and Stepping, ,Continuing and Stepping}) resumes execution until the
18725 selected stack frame returns naturally.
18727 @value{GDBN} needs to know how the @var{expression} argument should be set for
18728 the inferior. The concrete registers assignment depends on the OS ABI and the
18729 type being returned by the selected stack frame. For example it is common for
18730 OS ABI to return floating point values in FPU registers while integer values in
18731 CPU registers. Still some ABIs return even floating point values in CPU
18732 registers. Larger integer widths (such as @code{long long int}) also have
18733 specific placement rules. @value{GDBN} already knows the OS ABI from its
18734 current target so it needs to find out also the type being returned to make the
18735 assignment into the right register(s).
18737 Normally, the selected stack frame has debug info. @value{GDBN} will always
18738 use the debug info instead of the implicit type of @var{expression} when the
18739 debug info is available. For example, if you type @kbd{return -1}, and the
18740 function in the current stack frame is declared to return a @code{long long
18741 int}, @value{GDBN} transparently converts the implicit @code{int} value of -1
18742 into a @code{long long int}:
18745 Breakpoint 1, func () at gdb.base/return-nodebug.c:29
18747 (@value{GDBP}) return -1
18748 Make func return now? (y or n) y
18749 #0 0x004004f6 in main () at gdb.base/return-nodebug.c:43
18750 43 printf ("result=%lld\n", func ());
18754 However, if the selected stack frame does not have a debug info, e.g., if the
18755 function was compiled without debug info, @value{GDBN} has to find out the type
18756 to return from user. Specifying a different type by mistake may set the value
18757 in different inferior registers than the caller code expects. For example,
18758 typing @kbd{return -1} with its implicit type @code{int} would set only a part
18759 of a @code{long long int} result for a debug info less function (on 32-bit
18760 architectures). Therefore the user is required to specify the return type by
18761 an appropriate cast explicitly:
18764 Breakpoint 2, 0x0040050b in func ()
18765 (@value{GDBP}) return -1
18766 Return value type not available for selected stack frame.
18767 Please use an explicit cast of the value to return.
18768 (@value{GDBP}) return (long long int) -1
18769 Make selected stack frame return now? (y or n) y
18770 #0 0x00400526 in main ()
18775 @section Calling Program Functions
18778 @cindex calling functions
18779 @cindex inferior functions, calling
18780 @item print @var{expr}
18781 Evaluate the expression @var{expr} and display the resulting value.
18782 The expression may include calls to functions in the program being
18786 @item call @var{expr}
18787 Evaluate the expression @var{expr} without displaying @code{void}
18790 You can use this variant of the @code{print} command if you want to
18791 execute a function from your program that does not return anything
18792 (a.k.a.@: @dfn{a void function}), but without cluttering the output
18793 with @code{void} returned values that @value{GDBN} will otherwise
18794 print. If the result is not void, it is printed and saved in the
18798 It is possible for the function you call via the @code{print} or
18799 @code{call} command to generate a signal (e.g., if there's a bug in
18800 the function, or if you passed it incorrect arguments). What happens
18801 in that case is controlled by the @code{set unwindonsignal} command.
18803 Similarly, with a C@t{++} program it is possible for the function you
18804 call via the @code{print} or @code{call} command to generate an
18805 exception that is not handled due to the constraints of the dummy
18806 frame. In this case, any exception that is raised in the frame, but has
18807 an out-of-frame exception handler will not be found. GDB builds a
18808 dummy-frame for the inferior function call, and the unwinder cannot
18809 seek for exception handlers outside of this dummy-frame. What happens
18810 in that case is controlled by the
18811 @code{set unwind-on-terminating-exception} command.
18814 @item set unwindonsignal
18815 @kindex set unwindonsignal
18816 @cindex unwind stack in called functions
18817 @cindex call dummy stack unwinding
18818 Set unwinding of the stack if a signal is received while in a function
18819 that @value{GDBN} called in the program being debugged. If set to on,
18820 @value{GDBN} unwinds the stack it created for the call and restores
18821 the context to what it was before the call. If set to off (the
18822 default), @value{GDBN} stops in the frame where the signal was
18825 @item show unwindonsignal
18826 @kindex show unwindonsignal
18827 Show the current setting of stack unwinding in the functions called by
18830 @item set unwind-on-terminating-exception
18831 @kindex set unwind-on-terminating-exception
18832 @cindex unwind stack in called functions with unhandled exceptions
18833 @cindex call dummy stack unwinding on unhandled exception.
18834 Set unwinding of the stack if a C@t{++} exception is raised, but left
18835 unhandled while in a function that @value{GDBN} called in the program being
18836 debugged. If set to on (the default), @value{GDBN} unwinds the stack
18837 it created for the call and restores the context to what it was before
18838 the call. If set to off, @value{GDBN} the exception is delivered to
18839 the default C@t{++} exception handler and the inferior terminated.
18841 @item show unwind-on-terminating-exception
18842 @kindex show unwind-on-terminating-exception
18843 Show the current setting of stack unwinding in the functions called by
18846 @item set may-call-functions
18847 @kindex set may-call-functions
18848 @cindex disabling calling functions in the program
18849 @cindex calling functions in the program, disabling
18850 Set permission to call functions in the program.
18851 This controls whether @value{GDBN} will attempt to call functions in
18852 the program, such as with expressions in the @code{print} command. It
18853 defaults to @code{on}.
18855 To call a function in the program, @value{GDBN} has to temporarily
18856 modify the state of the inferior. This has potentially undesired side
18857 effects. Also, having @value{GDBN} call nested functions is likely to
18858 be erroneous and may even crash the program being debugged. You can
18859 avoid such hazards by forbidding @value{GDBN} from calling functions
18860 in the program being debugged. If calling functions in the program
18861 is forbidden, GDB will throw an error when a command (such as printing
18862 an expression) starts a function call in the program.
18864 @item show may-call-functions
18865 @kindex show may-call-functions
18866 Show permission to call functions in the program.
18870 @subsection Calling functions with no debug info
18872 @cindex no debug info functions
18873 Sometimes, a function you wish to call is missing debug information.
18874 In such case, @value{GDBN} does not know the type of the function,
18875 including the types of the function's parameters. To avoid calling
18876 the inferior function incorrectly, which could result in the called
18877 function functioning erroneously and even crash, @value{GDBN} refuses
18878 to call the function unless you tell it the type of the function.
18880 For prototyped (i.e.@: ANSI/ISO style) functions, there are two ways
18881 to do that. The simplest is to cast the call to the function's
18882 declared return type. For example:
18885 (@value{GDBP}) p getenv ("PATH")
18886 'getenv' has unknown return type; cast the call to its declared return type
18887 (@value{GDBP}) p (char *) getenv ("PATH")
18888 $1 = 0x7fffffffe7ba "/usr/local/bin:/"...
18891 Casting the return type of a no-debug function is equivalent to
18892 casting the function to a pointer to a prototyped function that has a
18893 prototype that matches the types of the passed-in arguments, and
18894 calling that. I.e., the call above is equivalent to:
18897 (@value{GDBP}) p ((char * (*) (const char *)) getenv) ("PATH")
18901 and given this prototyped C or C++ function with float parameters:
18904 float multiply (float v1, float v2) @{ return v1 * v2; @}
18908 these calls are equivalent:
18911 (@value{GDBP}) p (float) multiply (2.0f, 3.0f)
18912 (@value{GDBP}) p ((float (*) (float, float)) multiply) (2.0f, 3.0f)
18915 If the function you wish to call is declared as unprototyped (i.e.@:
18916 old K&R style), you must use the cast-to-function-pointer syntax, so
18917 that @value{GDBN} knows that it needs to apply default argument
18918 promotions (promote float arguments to double). @xref{ABI, float
18919 promotion}. For example, given this unprototyped C function with
18920 float parameters, and no debug info:
18924 multiply_noproto (v1, v2)
18932 you call it like this:
18935 (@value{GDBP}) p ((float (*) ()) multiply_noproto) (2.0f, 3.0f)
18939 @section Patching Programs
18941 @cindex patching binaries
18942 @cindex writing into executables
18943 @cindex writing into corefiles
18945 By default, @value{GDBN} opens the file containing your program's
18946 executable code (or the corefile) read-only. This prevents accidental
18947 alterations to machine code; but it also prevents you from intentionally
18948 patching your program's binary.
18950 If you'd like to be able to patch the binary, you can specify that
18951 explicitly with the @code{set write} command. For example, you might
18952 want to turn on internal debugging flags, or even to make emergency
18958 @itemx set write off
18959 If you specify @samp{set write on}, @value{GDBN} opens executable and
18960 core files for both reading and writing; if you specify @kbd{set write
18961 off} (the default), @value{GDBN} opens them read-only.
18963 If you have already loaded a file, you must load it again (using the
18964 @code{exec-file} or @code{core-file} command) after changing @code{set
18965 write}, for your new setting to take effect.
18969 Display whether executable files and core files are opened for writing
18970 as well as reading.
18973 @node Compiling and Injecting Code
18974 @section Compiling and injecting code in @value{GDBN}
18975 @cindex injecting code
18976 @cindex writing into executables
18977 @cindex compiling code
18979 @value{GDBN} supports on-demand compilation and code injection into
18980 programs running under @value{GDBN}. GCC 5.0 or higher built with
18981 @file{libcc1.so} must be installed for this functionality to be enabled.
18982 This functionality is implemented with the following commands.
18985 @kindex compile code
18986 @item compile code @var{source-code}
18987 @itemx compile code -raw @var{--} @var{source-code}
18988 Compile @var{source-code} with the compiler language found as the current
18989 language in @value{GDBN} (@pxref{Languages}). If compilation and
18990 injection is not supported with the current language specified in
18991 @value{GDBN}, or the compiler does not support this feature, an error
18992 message will be printed. If @var{source-code} compiles and links
18993 successfully, @value{GDBN} will load the object-code emitted,
18994 and execute it within the context of the currently selected inferior.
18995 It is important to note that the compiled code is executed immediately.
18996 After execution, the compiled code is removed from @value{GDBN} and any
18997 new types or variables you have defined will be deleted.
18999 The command allows you to specify @var{source-code} in two ways.
19000 The simplest method is to provide a single line of code to the command.
19004 compile code printf ("hello world\n");
19007 If you specify options on the command line as well as source code, they
19008 may conflict. The @samp{--} delimiter can be used to separate options
19009 from actual source code. E.g.:
19012 compile code -r -- printf ("hello world\n");
19015 Alternatively you can enter source code as multiple lines of text. To
19016 enter this mode, invoke the @samp{compile code} command without any text
19017 following the command. This will start the multiple-line editor and
19018 allow you to type as many lines of source code as required. When you
19019 have completed typing, enter @samp{end} on its own line to exit the
19024 >printf ("hello\n");
19025 >printf ("world\n");
19029 Specifying @samp{-raw}, prohibits @value{GDBN} from wrapping the
19030 provided @var{source-code} in a callable scope. In this case, you must
19031 specify the entry point of the code by defining a function named
19032 @code{_gdb_expr_}. The @samp{-raw} code cannot access variables of the
19033 inferior. Using @samp{-raw} option may be needed for example when
19034 @var{source-code} requires @samp{#include} lines which may conflict with
19035 inferior symbols otherwise.
19037 @kindex compile file
19038 @item compile file @var{filename}
19039 @itemx compile file -raw @var{filename}
19040 Like @code{compile code}, but take the source code from @var{filename}.
19043 compile file /home/user/example.c
19048 @item compile print @var{expr}
19049 @itemx compile print /@var{f} @var{expr}
19050 Compile and execute @var{expr} with the compiler language found as the
19051 current language in @value{GDBN} (@pxref{Languages}). By default the
19052 value of @var{expr} is printed in a format appropriate to its data type;
19053 you can choose a different format by specifying @samp{/@var{f}}, where
19054 @var{f} is a letter specifying the format; see @ref{Output Formats,,Output
19057 @item compile print
19058 @itemx compile print /@var{f}
19059 @cindex reprint the last value
19060 Alternatively you can enter the expression (source code producing it) as
19061 multiple lines of text. To enter this mode, invoke the @samp{compile print}
19062 command without any text following the command. This will start the
19063 multiple-line editor.
19067 The process of compiling and injecting the code can be inspected using:
19070 @anchor{set debug compile}
19071 @item set debug compile
19072 @cindex compile command debugging info
19073 Turns on or off display of @value{GDBN} process of compiling and
19074 injecting the code. The default is off.
19076 @item show debug compile
19077 Displays the current state of displaying @value{GDBN} process of
19078 compiling and injecting the code.
19080 @anchor{set debug compile-cplus-types}
19081 @item set debug compile-cplus-types
19082 @cindex compile C@t{++} type conversion
19083 Turns on or off the display of C@t{++} type conversion debugging information.
19084 The default is off.
19086 @item show debug compile-cplus-types
19087 Displays the current state of displaying debugging information for
19088 C@t{++} type conversion.
19091 @subsection Compilation options for the @code{compile} command
19093 @value{GDBN} needs to specify the right compilation options for the code
19094 to be injected, in part to make its ABI compatible with the inferior
19095 and in part to make the injected code compatible with @value{GDBN}'s
19099 The options used, in increasing precedence:
19102 @item target architecture and OS options (@code{gdbarch})
19103 These options depend on target processor type and target operating
19104 system, usually they specify at least 32-bit (@code{-m32}) or 64-bit
19105 (@code{-m64}) compilation option.
19107 @item compilation options recorded in the target
19108 @value{NGCC} (since version 4.7) stores the options used for compilation
19109 into @code{DW_AT_producer} part of DWARF debugging information according
19110 to the @value{NGCC} option @code{-grecord-gcc-switches}. One has to
19111 explicitly specify @code{-g} during inferior compilation otherwise
19112 @value{NGCC} produces no DWARF. This feature is only relevant for
19113 platforms where @code{-g} produces DWARF by default, otherwise one may
19114 try to enforce DWARF by using @code{-gdwarf-4}.
19116 @item compilation options set by @code{set compile-args}
19120 You can override compilation options using the following command:
19123 @item set compile-args
19124 @cindex compile command options override
19125 Set compilation options used for compiling and injecting code with the
19126 @code{compile} commands. These options override any conflicting ones
19127 from the target architecture and/or options stored during inferior
19130 @item show compile-args
19131 Displays the current state of compilation options override.
19132 This does not show all the options actually used during compilation,
19133 use @ref{set debug compile} for that.
19136 @subsection Caveats when using the @code{compile} command
19138 There are a few caveats to keep in mind when using the @code{compile}
19139 command. As the caveats are different per language, the table below
19140 highlights specific issues on a per language basis.
19143 @item C code examples and caveats
19144 When the language in @value{GDBN} is set to @samp{C}, the compiler will
19145 attempt to compile the source code with a @samp{C} compiler. The source
19146 code provided to the @code{compile} command will have much the same
19147 access to variables and types as it normally would if it were part of
19148 the program currently being debugged in @value{GDBN}.
19150 Below is a sample program that forms the basis of the examples that
19151 follow. This program has been compiled and loaded into @value{GDBN},
19152 much like any other normal debugging session.
19155 void function1 (void)
19158 printf ("function 1\n");
19161 void function2 (void)
19176 For the purposes of the examples in this section, the program above has
19177 been compiled, loaded into @value{GDBN}, stopped at the function
19178 @code{main}, and @value{GDBN} is awaiting input from the user.
19180 To access variables and types for any program in @value{GDBN}, the
19181 program must be compiled and packaged with debug information. The
19182 @code{compile} command is not an exception to this rule. Without debug
19183 information, you can still use the @code{compile} command, but you will
19184 be very limited in what variables and types you can access.
19186 So with that in mind, the example above has been compiled with debug
19187 information enabled. The @code{compile} command will have access to
19188 all variables and types (except those that may have been optimized
19189 out). Currently, as @value{GDBN} has stopped the program in the
19190 @code{main} function, the @code{compile} command would have access to
19191 the variable @code{k}. You could invoke the @code{compile} command
19192 and type some source code to set the value of @code{k}. You can also
19193 read it, or do anything with that variable you would normally do in
19194 @code{C}. Be aware that changes to inferior variables in the
19195 @code{compile} command are persistent. In the following example:
19198 compile code k = 3;
19202 the variable @code{k} is now 3. It will retain that value until
19203 something else in the example program changes it, or another
19204 @code{compile} command changes it.
19206 Normal scope and access rules apply to source code compiled and
19207 injected by the @code{compile} command. In the example, the variables
19208 @code{j} and @code{k} are not accessible yet, because the program is
19209 currently stopped in the @code{main} function, where these variables
19210 are not in scope. Therefore, the following command
19213 compile code j = 3;
19217 will result in a compilation error message.
19219 Once the program is continued, execution will bring these variables in
19220 scope, and they will become accessible; then the code you specify via
19221 the @code{compile} command will be able to access them.
19223 You can create variables and types with the @code{compile} command as
19224 part of your source code. Variables and types that are created as part
19225 of the @code{compile} command are not visible to the rest of the program for
19226 the duration of its run. This example is valid:
19229 compile code int ff = 5; printf ("ff is %d\n", ff);
19232 However, if you were to type the following into @value{GDBN} after that
19233 command has completed:
19236 compile code printf ("ff is %d\n'', ff);
19240 a compiler error would be raised as the variable @code{ff} no longer
19241 exists. Object code generated and injected by the @code{compile}
19242 command is removed when its execution ends. Caution is advised
19243 when assigning to program variables values of variables created by the
19244 code submitted to the @code{compile} command. This example is valid:
19247 compile code int ff = 5; k = ff;
19250 The value of the variable @code{ff} is assigned to @code{k}. The variable
19251 @code{k} does not require the existence of @code{ff} to maintain the value
19252 it has been assigned. However, pointers require particular care in
19253 assignment. If the source code compiled with the @code{compile} command
19254 changed the address of a pointer in the example program, perhaps to a
19255 variable created in the @code{compile} command, that pointer would point
19256 to an invalid location when the command exits. The following example
19257 would likely cause issues with your debugged program:
19260 compile code int ff = 5; p = &ff;
19263 In this example, @code{p} would point to @code{ff} when the
19264 @code{compile} command is executing the source code provided to it.
19265 However, as variables in the (example) program persist with their
19266 assigned values, the variable @code{p} would point to an invalid
19267 location when the command exists. A general rule should be followed
19268 in that you should either assign @code{NULL} to any assigned pointers,
19269 or restore a valid location to the pointer before the command exits.
19271 Similar caution must be exercised with any structs, unions, and typedefs
19272 defined in @code{compile} command. Types defined in the @code{compile}
19273 command will no longer be available in the next @code{compile} command.
19274 Therefore, if you cast a variable to a type defined in the
19275 @code{compile} command, care must be taken to ensure that any future
19276 need to resolve the type can be achieved.
19279 (gdb) compile code static struct a @{ int a; @} v = @{ 42 @}; argv = &v;
19280 (gdb) compile code printf ("%d\n", ((struct a *) argv)->a);
19281 gdb command line:1:36: error: dereferencing pointer to incomplete type ‘struct a’
19282 Compilation failed.
19283 (gdb) compile code struct a @{ int a; @}; printf ("%d\n", ((struct a *) argv)->a);
19287 Variables that have been optimized away by the compiler are not
19288 accessible to the code submitted to the @code{compile} command.
19289 Access to those variables will generate a compiler error which @value{GDBN}
19290 will print to the console.
19293 @subsection Compiler search for the @code{compile} command
19295 @value{GDBN} needs to find @value{NGCC} for the inferior being debugged
19296 which may not be obvious for remote targets of different architecture
19297 than where @value{GDBN} is running. Environment variable @code{PATH} on
19298 @value{GDBN} host is searched for @value{NGCC} binary matching the
19299 target architecture and operating system. This search can be overriden
19300 by @code{set compile-gcc} @value{GDBN} command below. @code{PATH} is
19301 taken from shell that executed @value{GDBN}, it is not the value set by
19302 @value{GDBN} command @code{set environment}). @xref{Environment}.
19305 Specifically @code{PATH} is searched for binaries matching regular expression
19306 @code{@var{arch}(-[^-]*)?-@var{os}-gcc} according to the inferior target being
19307 debugged. @var{arch} is processor name --- multiarch is supported, so for
19308 example both @code{i386} and @code{x86_64} targets look for pattern
19309 @code{(x86_64|i.86)} and both @code{s390} and @code{s390x} targets look
19310 for pattern @code{s390x?}. @var{os} is currently supported only for
19311 pattern @code{linux(-gnu)?}.
19313 On Posix hosts the compiler driver @value{GDBN} needs to find also
19314 shared library @file{libcc1.so} from the compiler. It is searched in
19315 default shared library search path (overridable with usual environment
19316 variable @code{LD_LIBRARY_PATH}), unrelated to @code{PATH} or @code{set
19317 compile-gcc} settings. Contrary to it @file{libcc1plugin.so} is found
19318 according to the installation of the found compiler --- as possibly
19319 specified by the @code{set compile-gcc} command.
19322 @item set compile-gcc
19323 @cindex compile command driver filename override
19324 Set compilation command used for compiling and injecting code with the
19325 @code{compile} commands. If this option is not set (it is set to
19326 an empty string), the search described above will occur --- that is the
19329 @item show compile-gcc
19330 Displays the current compile command @value{NGCC} driver filename.
19331 If set, it is the main command @command{gcc}, found usually for example
19332 under name @file{x86_64-linux-gnu-gcc}.
19336 @chapter @value{GDBN} Files
19338 @value{GDBN} needs to know the file name of the program to be debugged,
19339 both in order to read its symbol table and in order to start your
19340 program. To debug a core dump of a previous run, you must also tell
19341 @value{GDBN} the name of the core dump file.
19344 * Files:: Commands to specify files
19345 * File Caching:: Information about @value{GDBN}'s file caching
19346 * Separate Debug Files:: Debugging information in separate files
19347 * MiniDebugInfo:: Debugging information in a special section
19348 * Index Files:: Index files speed up GDB
19349 * Symbol Errors:: Errors reading symbol files
19350 * Data Files:: GDB data files
19354 @section Commands to Specify Files
19356 @cindex symbol table
19357 @cindex core dump file
19359 You may want to specify executable and core dump file names. The usual
19360 way to do this is at start-up time, using the arguments to
19361 @value{GDBN}'s start-up commands (@pxref{Invocation, , Getting In and
19362 Out of @value{GDBN}}).
19364 Occasionally it is necessary to change to a different file during a
19365 @value{GDBN} session. Or you may run @value{GDBN} and forget to
19366 specify a file you want to use. Or you are debugging a remote target
19367 via @code{gdbserver} (@pxref{Server, file, Using the @code{gdbserver}
19368 Program}). In these situations the @value{GDBN} commands to specify
19369 new files are useful.
19372 @cindex executable file
19374 @item file @var{filename}
19375 Use @var{filename} as the program to be debugged. It is read for its
19376 symbols and for the contents of pure memory. It is also the program
19377 executed when you use the @code{run} command. If you do not specify a
19378 directory and the file is not found in the @value{GDBN} working directory,
19379 @value{GDBN} uses the environment variable @code{PATH} as a list of
19380 directories to search, just as the shell does when looking for a program
19381 to run. You can change the value of this variable, for both @value{GDBN}
19382 and your program, using the @code{path} command.
19384 @cindex unlinked object files
19385 @cindex patching object files
19386 You can load unlinked object @file{.o} files into @value{GDBN} using
19387 the @code{file} command. You will not be able to ``run'' an object
19388 file, but you can disassemble functions and inspect variables. Also,
19389 if the underlying BFD functionality supports it, you could use
19390 @kbd{gdb -write} to patch object files using this technique. Note
19391 that @value{GDBN} can neither interpret nor modify relocations in this
19392 case, so branches and some initialized variables will appear to go to
19393 the wrong place. But this feature is still handy from time to time.
19396 @code{file} with no argument makes @value{GDBN} discard any information it
19397 has on both executable file and the symbol table.
19400 @item exec-file @r{[} @var{filename} @r{]}
19401 Specify that the program to be run (but not the symbol table) is found
19402 in @var{filename}. @value{GDBN} searches the environment variable @code{PATH}
19403 if necessary to locate your program. Omitting @var{filename} means to
19404 discard information on the executable file.
19406 @kindex symbol-file
19407 @item symbol-file @r{[} @var{filename} @r{[} -o @var{offset} @r{]]}
19408 Read symbol table information from file @var{filename}. @code{PATH} is
19409 searched when necessary. Use the @code{file} command to get both symbol
19410 table and program to run from the same file.
19412 If an optional @var{offset} is specified, it is added to the start
19413 address of each section in the symbol file. This is useful if the
19414 program is relocated at runtime, such as the Linux kernel with kASLR
19417 @code{symbol-file} with no argument clears out @value{GDBN} information on your
19418 program's symbol table.
19420 The @code{symbol-file} command causes @value{GDBN} to forget the contents of
19421 some breakpoints and auto-display expressions. This is because they may
19422 contain pointers to the internal data recording symbols and data types,
19423 which are part of the old symbol table data being discarded inside
19426 @code{symbol-file} does not repeat if you press @key{RET} again after
19429 When @value{GDBN} is configured for a particular environment, it
19430 understands debugging information in whatever format is the standard
19431 generated for that environment; you may use either a @sc{gnu} compiler, or
19432 other compilers that adhere to the local conventions.
19433 Best results are usually obtained from @sc{gnu} compilers; for example,
19434 using @code{@value{NGCC}} you can generate debugging information for
19437 For most kinds of object files, with the exception of old SVR3 systems
19438 using COFF, the @code{symbol-file} command does not normally read the
19439 symbol table in full right away. Instead, it scans the symbol table
19440 quickly to find which source files and which symbols are present. The
19441 details are read later, one source file at a time, as they are needed.
19443 The purpose of this two-stage reading strategy is to make @value{GDBN}
19444 start up faster. For the most part, it is invisible except for
19445 occasional pauses while the symbol table details for a particular source
19446 file are being read. (The @code{set verbose} command can turn these
19447 pauses into messages if desired. @xref{Messages/Warnings, ,Optional
19448 Warnings and Messages}.)
19450 We have not implemented the two-stage strategy for COFF yet. When the
19451 symbol table is stored in COFF format, @code{symbol-file} reads the
19452 symbol table data in full right away. Note that ``stabs-in-COFF''
19453 still does the two-stage strategy, since the debug info is actually
19457 @cindex reading symbols immediately
19458 @cindex symbols, reading immediately
19459 @item symbol-file @r{[} -readnow @r{]} @var{filename}
19460 @itemx file @r{[} -readnow @r{]} @var{filename}
19461 You can override the @value{GDBN} two-stage strategy for reading symbol
19462 tables by using the @samp{-readnow} option with any of the commands that
19463 load symbol table information, if you want to be sure @value{GDBN} has the
19464 entire symbol table available.
19466 @cindex @code{-readnever}, option for symbol-file command
19467 @cindex never read symbols
19468 @cindex symbols, never read
19469 @item symbol-file @r{[} -readnever @r{]} @var{filename}
19470 @itemx file @r{[} -readnever @r{]} @var{filename}
19471 You can instruct @value{GDBN} to never read the symbolic information
19472 contained in @var{filename} by using the @samp{-readnever} option.
19473 @xref{--readnever}.
19475 @c FIXME: for now no mention of directories, since this seems to be in
19476 @c flux. 13mar1992 status is that in theory GDB would look either in
19477 @c current dir or in same dir as myprog; but issues like competing
19478 @c GDB's, or clutter in system dirs, mean that in practice right now
19479 @c only current dir is used. FFish says maybe a special GDB hierarchy
19480 @c (eg rooted in val of env var GDBSYMS) could exist for mappable symbol
19484 @item core-file @r{[}@var{filename}@r{]}
19486 Specify the whereabouts of a core dump file to be used as the ``contents
19487 of memory''. Traditionally, core files contain only some parts of the
19488 address space of the process that generated them; @value{GDBN} can access the
19489 executable file itself for other parts.
19491 @code{core-file} with no argument specifies that no core file is
19494 Note that the core file is ignored when your program is actually running
19495 under @value{GDBN}. So, if you have been running your program and you
19496 wish to debug a core file instead, you must kill the subprocess in which
19497 the program is running. To do this, use the @code{kill} command
19498 (@pxref{Kill Process, ,Killing the Child Process}).
19500 @kindex add-symbol-file
19501 @cindex dynamic linking
19502 @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{]}
19503 The @code{add-symbol-file} command reads additional symbol table
19504 information from the file @var{filename}. You would use this command
19505 when @var{filename} has been dynamically loaded (by some other means)
19506 into the program that is running. The @var{textaddress} parameter gives
19507 the memory address at which the file's text section has been loaded.
19508 You can additionally specify the base address of other sections using
19509 an arbitrary number of @samp{-s @var{section} @var{address}} pairs.
19510 If a section is omitted, @value{GDBN} will use its default addresses
19511 as found in @var{filename}. Any @var{address} or @var{textaddress}
19512 can be given as an expression.
19514 If an optional @var{offset} is specified, it is added to the start
19515 address of each section, except those for which the address was
19516 specified explicitly.
19518 The symbol table of the file @var{filename} is added to the symbol table
19519 originally read with the @code{symbol-file} command. You can use the
19520 @code{add-symbol-file} command any number of times; the new symbol data
19521 thus read is kept in addition to the old.
19523 Changes can be reverted using the command @code{remove-symbol-file}.
19525 @cindex relocatable object files, reading symbols from
19526 @cindex object files, relocatable, reading symbols from
19527 @cindex reading symbols from relocatable object files
19528 @cindex symbols, reading from relocatable object files
19529 @cindex @file{.o} files, reading symbols from
19530 Although @var{filename} is typically a shared library file, an
19531 executable file, or some other object file which has been fully
19532 relocated for loading into a process, you can also load symbolic
19533 information from relocatable @file{.o} files, as long as:
19537 the file's symbolic information refers only to linker symbols defined in
19538 that file, not to symbols defined by other object files,
19540 every section the file's symbolic information refers to has actually
19541 been loaded into the inferior, as it appears in the file, and
19543 you can determine the address at which every section was loaded, and
19544 provide these to the @code{add-symbol-file} command.
19548 Some embedded operating systems, like Sun Chorus and VxWorks, can load
19549 relocatable files into an already running program; such systems
19550 typically make the requirements above easy to meet. However, it's
19551 important to recognize that many native systems use complex link
19552 procedures (@code{.linkonce} section factoring and C@t{++} constructor table
19553 assembly, for example) that make the requirements difficult to meet. In
19554 general, one cannot assume that using @code{add-symbol-file} to read a
19555 relocatable object file's symbolic information will have the same effect
19556 as linking the relocatable object file into the program in the normal
19559 @code{add-symbol-file} does not repeat if you press @key{RET} after using it.
19561 @kindex remove-symbol-file
19562 @item remove-symbol-file @var{filename}
19563 @item remove-symbol-file -a @var{address}
19564 Remove a symbol file added via the @code{add-symbol-file} command. The
19565 file to remove can be identified by its @var{filename} or by an @var{address}
19566 that lies within the boundaries of this symbol file in memory. Example:
19569 (gdb) add-symbol-file /home/user/gdb/mylib.so 0x7ffff7ff9480
19570 add symbol table from file "/home/user/gdb/mylib.so" at
19571 .text_addr = 0x7ffff7ff9480
19573 Reading symbols from /home/user/gdb/mylib.so...done.
19574 (gdb) remove-symbol-file -a 0x7ffff7ff9480
19575 Remove symbol table from file "/home/user/gdb/mylib.so"? (y or n) y
19580 @code{remove-symbol-file} does not repeat if you press @key{RET} after using it.
19582 @kindex add-symbol-file-from-memory
19583 @cindex @code{syscall DSO}
19584 @cindex load symbols from memory
19585 @item add-symbol-file-from-memory @var{address}
19586 Load symbols from the given @var{address} in a dynamically loaded
19587 object file whose image is mapped directly into the inferior's memory.
19588 For example, the Linux kernel maps a @code{syscall DSO} into each
19589 process's address space; this DSO provides kernel-specific code for
19590 some system calls. The argument can be any expression whose
19591 evaluation yields the address of the file's shared object file header.
19592 For this command to work, you must have used @code{symbol-file} or
19593 @code{exec-file} commands in advance.
19596 @item section @var{section} @var{addr}
19597 The @code{section} command changes the base address of the named
19598 @var{section} of the exec file to @var{addr}. This can be used if the
19599 exec file does not contain section addresses, (such as in the
19600 @code{a.out} format), or when the addresses specified in the file
19601 itself are wrong. Each section must be changed separately. The
19602 @code{info files} command, described below, lists all the sections and
19606 @kindex info target
19609 @code{info files} and @code{info target} are synonymous; both print the
19610 current target (@pxref{Targets, ,Specifying a Debugging Target}),
19611 including the names of the executable and core dump files currently in
19612 use by @value{GDBN}, and the files from which symbols were loaded. The
19613 command @code{help target} lists all possible targets rather than
19616 @kindex maint info sections
19617 @item maint info sections
19618 Another command that can give you extra information about program sections
19619 is @code{maint info sections}. In addition to the section information
19620 displayed by @code{info files}, this command displays the flags and file
19621 offset of each section in the executable and core dump files. In addition,
19622 @code{maint info sections} provides the following command options (which
19623 may be arbitrarily combined):
19627 Display sections for all loaded object files, including shared libraries.
19628 @item @var{sections}
19629 Display info only for named @var{sections}.
19630 @item @var{section-flags}
19631 Display info only for sections for which @var{section-flags} are true.
19632 The section flags that @value{GDBN} currently knows about are:
19635 Section will have space allocated in the process when loaded.
19636 Set for all sections except those containing debug information.
19638 Section will be loaded from the file into the child process memory.
19639 Set for pre-initialized code and data, clear for @code{.bss} sections.
19641 Section needs to be relocated before loading.
19643 Section cannot be modified by the child process.
19645 Section contains executable code only.
19647 Section contains data only (no executable code).
19649 Section will reside in ROM.
19651 Section contains data for constructor/destructor lists.
19653 Section is not empty.
19655 An instruction to the linker to not output the section.
19656 @item COFF_SHARED_LIBRARY
19657 A notification to the linker that the section contains
19658 COFF shared library information.
19660 Section contains common symbols.
19663 @kindex set trust-readonly-sections
19664 @cindex read-only sections
19665 @item set trust-readonly-sections on
19666 Tell @value{GDBN} that readonly sections in your object file
19667 really are read-only (i.e.@: that their contents will not change).
19668 In that case, @value{GDBN} can fetch values from these sections
19669 out of the object file, rather than from the target program.
19670 For some targets (notably embedded ones), this can be a significant
19671 enhancement to debugging performance.
19673 The default is off.
19675 @item set trust-readonly-sections off
19676 Tell @value{GDBN} not to trust readonly sections. This means that
19677 the contents of the section might change while the program is running,
19678 and must therefore be fetched from the target when needed.
19680 @item show trust-readonly-sections
19681 Show the current setting of trusting readonly sections.
19684 All file-specifying commands allow both absolute and relative file names
19685 as arguments. @value{GDBN} always converts the file name to an absolute file
19686 name and remembers it that way.
19688 @cindex shared libraries
19689 @anchor{Shared Libraries}
19690 @value{GDBN} supports @sc{gnu}/Linux, MS-Windows, SunOS,
19691 Darwin/Mach-O, SVr4, IBM RS/6000 AIX, QNX Neutrino, FDPIC (FR-V), and
19692 DSBT (TIC6X) shared libraries.
19694 On MS-Windows @value{GDBN} must be linked with the Expat library to support
19695 shared libraries. @xref{Expat}.
19697 @value{GDBN} automatically loads symbol definitions from shared libraries
19698 when you use the @code{run} command, or when you examine a core file.
19699 (Before you issue the @code{run} command, @value{GDBN} does not understand
19700 references to a function in a shared library, however---unless you are
19701 debugging a core file).
19703 @c FIXME: some @value{GDBN} release may permit some refs to undef
19704 @c FIXME...symbols---eg in a break cmd---assuming they are from a shared
19705 @c FIXME...lib; check this from time to time when updating manual
19707 There are times, however, when you may wish to not automatically load
19708 symbol definitions from shared libraries, such as when they are
19709 particularly large or there are many of them.
19711 To control the automatic loading of shared library symbols, use the
19715 @kindex set auto-solib-add
19716 @item set auto-solib-add @var{mode}
19717 If @var{mode} is @code{on}, symbols from all shared object libraries
19718 will be loaded automatically when the inferior begins execution, you
19719 attach to an independently started inferior, or when the dynamic linker
19720 informs @value{GDBN} that a new library has been loaded. If @var{mode}
19721 is @code{off}, symbols must be loaded manually, using the
19722 @code{sharedlibrary} command. The default value is @code{on}.
19724 @cindex memory used for symbol tables
19725 If your program uses lots of shared libraries with debug info that
19726 takes large amounts of memory, you can decrease the @value{GDBN}
19727 memory footprint by preventing it from automatically loading the
19728 symbols from shared libraries. To that end, type @kbd{set
19729 auto-solib-add off} before running the inferior, then load each
19730 library whose debug symbols you do need with @kbd{sharedlibrary
19731 @var{regexp}}, where @var{regexp} is a regular expression that matches
19732 the libraries whose symbols you want to be loaded.
19734 @kindex show auto-solib-add
19735 @item show auto-solib-add
19736 Display the current autoloading mode.
19739 @cindex load shared library
19740 To explicitly load shared library symbols, use the @code{sharedlibrary}
19744 @kindex info sharedlibrary
19746 @item info share @var{regex}
19747 @itemx info sharedlibrary @var{regex}
19748 Print the names of the shared libraries which are currently loaded
19749 that match @var{regex}. If @var{regex} is omitted then print
19750 all shared libraries that are loaded.
19753 @item info dll @var{regex}
19754 This is an alias of @code{info sharedlibrary}.
19756 @kindex sharedlibrary
19758 @item sharedlibrary @var{regex}
19759 @itemx share @var{regex}
19760 Load shared object library symbols for files matching a
19761 Unix regular expression.
19762 As with files loaded automatically, it only loads shared libraries
19763 required by your program for a core file or after typing @code{run}. If
19764 @var{regex} is omitted all shared libraries required by your program are
19767 @item nosharedlibrary
19768 @kindex nosharedlibrary
19769 @cindex unload symbols from shared libraries
19770 Unload all shared object library symbols. This discards all symbols
19771 that have been loaded from all shared libraries. Symbols from shared
19772 libraries that were loaded by explicit user requests are not
19776 Sometimes you may wish that @value{GDBN} stops and gives you control
19777 when any of shared library events happen. The best way to do this is
19778 to use @code{catch load} and @code{catch unload} (@pxref{Set
19781 @value{GDBN} also supports the the @code{set stop-on-solib-events}
19782 command for this. This command exists for historical reasons. It is
19783 less useful than setting a catchpoint, because it does not allow for
19784 conditions or commands as a catchpoint does.
19787 @item set stop-on-solib-events
19788 @kindex set stop-on-solib-events
19789 This command controls whether @value{GDBN} should give you control
19790 when the dynamic linker notifies it about some shared library event.
19791 The most common event of interest is loading or unloading of a new
19794 @item show stop-on-solib-events
19795 @kindex show stop-on-solib-events
19796 Show whether @value{GDBN} stops and gives you control when shared
19797 library events happen.
19800 Shared libraries are also supported in many cross or remote debugging
19801 configurations. @value{GDBN} needs to have access to the target's libraries;
19802 this can be accomplished either by providing copies of the libraries
19803 on the host system, or by asking @value{GDBN} to automatically retrieve the
19804 libraries from the target. If copies of the target libraries are
19805 provided, they need to be the same as the target libraries, although the
19806 copies on the target can be stripped as long as the copies on the host are
19809 @cindex where to look for shared libraries
19810 For remote debugging, you need to tell @value{GDBN} where the target
19811 libraries are, so that it can load the correct copies---otherwise, it
19812 may try to load the host's libraries. @value{GDBN} has two variables
19813 to specify the search directories for target libraries.
19816 @cindex prefix for executable and shared library file names
19817 @cindex system root, alternate
19818 @kindex set solib-absolute-prefix
19819 @kindex set sysroot
19820 @item set sysroot @var{path}
19821 Use @var{path} as the system root for the program being debugged. Any
19822 absolute shared library paths will be prefixed with @var{path}; many
19823 runtime loaders store the absolute paths to the shared library in the
19824 target program's memory. When starting processes remotely, and when
19825 attaching to already-running processes (local or remote), their
19826 executable filenames will be prefixed with @var{path} if reported to
19827 @value{GDBN} as absolute by the operating system. If you use
19828 @code{set sysroot} to find executables and shared libraries, they need
19829 to be laid out in the same way that they are on the target, with
19830 e.g.@: a @file{/bin}, @file{/lib} and @file{/usr/lib} hierarchy under
19833 If @var{path} starts with the sequence @file{target:} and the target
19834 system is remote then @value{GDBN} will retrieve the target binaries
19835 from the remote system. This is only supported when using a remote
19836 target that supports the @code{remote get} command (@pxref{File
19837 Transfer,,Sending files to a remote system}). The part of @var{path}
19838 following the initial @file{target:} (if present) is used as system
19839 root prefix on the remote file system. If @var{path} starts with the
19840 sequence @file{remote:} this is converted to the sequence
19841 @file{target:} by @code{set sysroot}@footnote{Historically the
19842 functionality to retrieve binaries from the remote system was
19843 provided by prefixing @var{path} with @file{remote:}}. If you want
19844 to specify a local system root using a directory that happens to be
19845 named @file{target:} or @file{remote:}, you need to use some
19846 equivalent variant of the name like @file{./target:}.
19848 For targets with an MS-DOS based filesystem, such as MS-Windows and
19849 SymbianOS, @value{GDBN} tries prefixing a few variants of the target
19850 absolute file name with @var{path}. But first, on Unix hosts,
19851 @value{GDBN} converts all backslash directory separators into forward
19852 slashes, because the backslash is not a directory separator on Unix:
19855 c:\foo\bar.dll @result{} c:/foo/bar.dll
19858 Then, @value{GDBN} attempts prefixing the target file name with
19859 @var{path}, and looks for the resulting file name in the host file
19863 c:/foo/bar.dll @result{} /path/to/sysroot/c:/foo/bar.dll
19866 If that does not find the binary, @value{GDBN} tries removing
19867 the @samp{:} character from the drive spec, both for convenience, and,
19868 for the case of the host file system not supporting file names with
19872 c:/foo/bar.dll @result{} /path/to/sysroot/c/foo/bar.dll
19875 This makes it possible to have a system root that mirrors a target
19876 with more than one drive. E.g., you may want to setup your local
19877 copies of the target system shared libraries like so (note @samp{c} vs
19881 @file{/path/to/sysroot/c/sys/bin/foo.dll}
19882 @file{/path/to/sysroot/c/sys/bin/bar.dll}
19883 @file{/path/to/sysroot/z/sys/bin/bar.dll}
19887 and point the system root at @file{/path/to/sysroot}, so that
19888 @value{GDBN} can find the correct copies of both
19889 @file{c:\sys\bin\foo.dll}, and @file{z:\sys\bin\bar.dll}.
19891 If that still does not find the binary, @value{GDBN} tries
19892 removing the whole drive spec from the target file name:
19895 c:/foo/bar.dll @result{} /path/to/sysroot/foo/bar.dll
19898 This last lookup makes it possible to not care about the drive name,
19899 if you don't want or need to.
19901 The @code{set solib-absolute-prefix} command is an alias for @code{set
19904 @cindex default system root
19905 @cindex @samp{--with-sysroot}
19906 You can set the default system root by using the configure-time
19907 @samp{--with-sysroot} option. If the system root is inside
19908 @value{GDBN}'s configured binary prefix (set with @samp{--prefix} or
19909 @samp{--exec-prefix}), then the default system root will be updated
19910 automatically if the installed @value{GDBN} is moved to a new
19913 @kindex show sysroot
19915 Display the current executable and shared library prefix.
19917 @kindex set solib-search-path
19918 @item set solib-search-path @var{path}
19919 If this variable is set, @var{path} is a colon-separated list of
19920 directories to search for shared libraries. @samp{solib-search-path}
19921 is used after @samp{sysroot} fails to locate the library, or if the
19922 path to the library is relative instead of absolute. If you want to
19923 use @samp{solib-search-path} instead of @samp{sysroot}, be sure to set
19924 @samp{sysroot} to a nonexistent directory to prevent @value{GDBN} from
19925 finding your host's libraries. @samp{sysroot} is preferred; setting
19926 it to a nonexistent directory may interfere with automatic loading
19927 of shared library symbols.
19929 @kindex show solib-search-path
19930 @item show solib-search-path
19931 Display the current shared library search path.
19933 @cindex DOS file-name semantics of file names.
19934 @kindex set target-file-system-kind (unix|dos-based|auto)
19935 @kindex show target-file-system-kind
19936 @item set target-file-system-kind @var{kind}
19937 Set assumed file system kind for target reported file names.
19939 Shared library file names as reported by the target system may not
19940 make sense as is on the system @value{GDBN} is running on. For
19941 example, when remote debugging a target that has MS-DOS based file
19942 system semantics, from a Unix host, the target may be reporting to
19943 @value{GDBN} a list of loaded shared libraries with file names such as
19944 @file{c:\Windows\kernel32.dll}. On Unix hosts, there's no concept of
19945 drive letters, so the @samp{c:\} prefix is not normally understood as
19946 indicating an absolute file name, and neither is the backslash
19947 normally considered a directory separator character. In that case,
19948 the native file system would interpret this whole absolute file name
19949 as a relative file name with no directory components. This would make
19950 it impossible to point @value{GDBN} at a copy of the remote target's
19951 shared libraries on the host using @code{set sysroot}, and impractical
19952 with @code{set solib-search-path}. Setting
19953 @code{target-file-system-kind} to @code{dos-based} tells @value{GDBN}
19954 to interpret such file names similarly to how the target would, and to
19955 map them to file names valid on @value{GDBN}'s native file system
19956 semantics. The value of @var{kind} can be @code{"auto"}, in addition
19957 to one of the supported file system kinds. In that case, @value{GDBN}
19958 tries to determine the appropriate file system variant based on the
19959 current target's operating system (@pxref{ABI, ,Configuring the
19960 Current ABI}). The supported file system settings are:
19964 Instruct @value{GDBN} to assume the target file system is of Unix
19965 kind. Only file names starting the forward slash (@samp{/}) character
19966 are considered absolute, and the directory separator character is also
19970 Instruct @value{GDBN} to assume the target file system is DOS based.
19971 File names starting with either a forward slash, or a drive letter
19972 followed by a colon (e.g., @samp{c:}), are considered absolute, and
19973 both the slash (@samp{/}) and the backslash (@samp{\\}) characters are
19974 considered directory separators.
19977 Instruct @value{GDBN} to use the file system kind associated with the
19978 target operating system (@pxref{ABI, ,Configuring the Current ABI}).
19979 This is the default.
19983 @cindex file name canonicalization
19984 @cindex base name differences
19985 When processing file names provided by the user, @value{GDBN}
19986 frequently needs to compare them to the file names recorded in the
19987 program's debug info. Normally, @value{GDBN} compares just the
19988 @dfn{base names} of the files as strings, which is reasonably fast
19989 even for very large programs. (The base name of a file is the last
19990 portion of its name, after stripping all the leading directories.)
19991 This shortcut in comparison is based upon the assumption that files
19992 cannot have more than one base name. This is usually true, but
19993 references to files that use symlinks or similar filesystem
19994 facilities violate that assumption. If your program records files
19995 using such facilities, or if you provide file names to @value{GDBN}
19996 using symlinks etc., you can set @code{basenames-may-differ} to
19997 @code{true} to instruct @value{GDBN} to completely canonicalize each
19998 pair of file names it needs to compare. This will make file-name
19999 comparisons accurate, but at a price of a significant slowdown.
20002 @item set basenames-may-differ
20003 @kindex set basenames-may-differ
20004 Set whether a source file may have multiple base names.
20006 @item show basenames-may-differ
20007 @kindex show basenames-may-differ
20008 Show whether a source file may have multiple base names.
20012 @section File Caching
20013 @cindex caching of opened files
20014 @cindex caching of bfd objects
20016 To speed up file loading, and reduce memory usage, @value{GDBN} will
20017 reuse the @code{bfd} objects used to track open files. @xref{Top, ,
20018 BFD, bfd, The Binary File Descriptor Library}. The following commands
20019 allow visibility and control of the caching behavior.
20022 @kindex maint info bfds
20023 @item maint info bfds
20024 This prints information about each @code{bfd} object that is known to
20027 @kindex maint set bfd-sharing
20028 @kindex maint show bfd-sharing
20029 @kindex bfd caching
20030 @item maint set bfd-sharing
20031 @item maint show bfd-sharing
20032 Control whether @code{bfd} objects can be shared. When sharing is
20033 enabled @value{GDBN} reuses already open @code{bfd} objects rather
20034 than reopening the same file. Turning sharing off does not cause
20035 already shared @code{bfd} objects to be unshared, but all future files
20036 that are opened will create a new @code{bfd} object. Similarly,
20037 re-enabling sharing does not cause multiple existing @code{bfd}
20038 objects to be collapsed into a single shared @code{bfd} object.
20040 @kindex set debug bfd-cache @var{level}
20041 @kindex bfd caching
20042 @item set debug bfd-cache @var{level}
20043 Turns on debugging of the bfd cache, setting the level to @var{level}.
20045 @kindex show debug bfd-cache
20046 @kindex bfd caching
20047 @item show debug bfd-cache
20048 Show the current debugging level of the bfd cache.
20051 @node Separate Debug Files
20052 @section Debugging Information in Separate Files
20053 @cindex separate debugging information files
20054 @cindex debugging information in separate files
20055 @cindex @file{.debug} subdirectories
20056 @cindex debugging information directory, global
20057 @cindex global debugging information directories
20058 @cindex build ID, and separate debugging files
20059 @cindex @file{.build-id} directory
20061 @value{GDBN} allows you to put a program's debugging information in a
20062 file separate from the executable itself, in a way that allows
20063 @value{GDBN} to find and load the debugging information automatically.
20064 Since debugging information can be very large---sometimes larger
20065 than the executable code itself---some systems distribute debugging
20066 information for their executables in separate files, which users can
20067 install only when they need to debug a problem.
20069 @value{GDBN} supports two ways of specifying the separate debug info
20074 The executable contains a @dfn{debug link} that specifies the name of
20075 the separate debug info file. The separate debug file's name is
20076 usually @file{@var{executable}.debug}, where @var{executable} is the
20077 name of the corresponding executable file without leading directories
20078 (e.g., @file{ls.debug} for @file{/usr/bin/ls}). In addition, the
20079 debug link specifies a 32-bit @dfn{Cyclic Redundancy Check} (CRC)
20080 checksum for the debug file, which @value{GDBN} uses to validate that
20081 the executable and the debug file came from the same build.
20084 The executable contains a @dfn{build ID}, a unique bit string that is
20085 also present in the corresponding debug info file. (This is supported
20086 only on some operating systems, when using the ELF or PE file formats
20087 for binary files and the @sc{gnu} Binutils.) For more details about
20088 this feature, see the description of the @option{--build-id}
20089 command-line option in @ref{Options, , Command Line Options, ld,
20090 The GNU Linker}. The debug info file's name is not specified
20091 explicitly by the build ID, but can be computed from the build ID, see
20095 Depending on the way the debug info file is specified, @value{GDBN}
20096 uses two different methods of looking for the debug file:
20100 For the ``debug link'' method, @value{GDBN} looks up the named file in
20101 the directory of the executable file, then in a subdirectory of that
20102 directory named @file{.debug}, and finally under each one of the
20103 global debug directories, in a subdirectory whose name is identical to
20104 the leading directories of the executable's absolute file name. (On
20105 MS-Windows/MS-DOS, the drive letter of the executable's leading
20106 directories is converted to a one-letter subdirectory, i.e.@:
20107 @file{d:/usr/bin/} is converted to @file{/d/usr/bin/}, because Windows
20108 filesystems disallow colons in file names.)
20111 For the ``build ID'' method, @value{GDBN} looks in the
20112 @file{.build-id} subdirectory of each one of the global debug directories for
20113 a file named @file{@var{nn}/@var{nnnnnnnn}.debug}, where @var{nn} are the
20114 first 2 hex characters of the build ID bit string, and @var{nnnnnnnn}
20115 are the rest of the bit string. (Real build ID strings are 32 or more
20116 hex characters, not 10.)
20119 So, for example, suppose you ask @value{GDBN} to debug
20120 @file{/usr/bin/ls}, which has a debug link that specifies the
20121 file @file{ls.debug}, and a build ID whose value in hex is
20122 @code{abcdef1234}. If the list of the global debug directories includes
20123 @file{/usr/lib/debug}, then @value{GDBN} will look for the following
20124 debug information files, in the indicated order:
20128 @file{/usr/lib/debug/.build-id/ab/cdef1234.debug}
20130 @file{/usr/bin/ls.debug}
20132 @file{/usr/bin/.debug/ls.debug}
20134 @file{/usr/lib/debug/usr/bin/ls.debug}.
20137 @anchor{debug-file-directory}
20138 Global debugging info directories default to what is set by @value{GDBN}
20139 configure option @option{--with-separate-debug-dir}. During @value{GDBN} run
20140 you can also set the global debugging info directories, and view the list
20141 @value{GDBN} is currently using.
20145 @kindex set debug-file-directory
20146 @item set debug-file-directory @var{directories}
20147 Set the directories which @value{GDBN} searches for separate debugging
20148 information files to @var{directory}. Multiple path components can be set
20149 concatenating them by a path separator.
20151 @kindex show debug-file-directory
20152 @item show debug-file-directory
20153 Show the directories @value{GDBN} searches for separate debugging
20158 @cindex @code{.gnu_debuglink} sections
20159 @cindex debug link sections
20160 A debug link is a special section of the executable file named
20161 @code{.gnu_debuglink}. The section must contain:
20165 A filename, with any leading directory components removed, followed by
20168 zero to three bytes of padding, as needed to reach the next four-byte
20169 boundary within the section, and
20171 a four-byte CRC checksum, stored in the same endianness used for the
20172 executable file itself. The checksum is computed on the debugging
20173 information file's full contents by the function given below, passing
20174 zero as the @var{crc} argument.
20177 Any executable file format can carry a debug link, as long as it can
20178 contain a section named @code{.gnu_debuglink} with the contents
20181 @cindex @code{.note.gnu.build-id} sections
20182 @cindex build ID sections
20183 The build ID is a special section in the executable file (and in other
20184 ELF binary files that @value{GDBN} may consider). This section is
20185 often named @code{.note.gnu.build-id}, but that name is not mandatory.
20186 It contains unique identification for the built files---the ID remains
20187 the same across multiple builds of the same build tree. The default
20188 algorithm SHA1 produces 160 bits (40 hexadecimal characters) of the
20189 content for the build ID string. The same section with an identical
20190 value is present in the original built binary with symbols, in its
20191 stripped variant, and in the separate debugging information file.
20193 The debugging information file itself should be an ordinary
20194 executable, containing a full set of linker symbols, sections, and
20195 debugging information. The sections of the debugging information file
20196 should have the same names, addresses, and sizes as the original file,
20197 but they need not contain any data---much like a @code{.bss} section
20198 in an ordinary executable.
20200 The @sc{gnu} binary utilities (Binutils) package includes the
20201 @samp{objcopy} utility that can produce
20202 the separated executable / debugging information file pairs using the
20203 following commands:
20206 @kbd{objcopy --only-keep-debug foo foo.debug}
20211 These commands remove the debugging
20212 information from the executable file @file{foo} and place it in the file
20213 @file{foo.debug}. You can use the first, second or both methods to link the
20218 The debug link method needs the following additional command to also leave
20219 behind a debug link in @file{foo}:
20222 @kbd{objcopy --add-gnu-debuglink=foo.debug foo}
20225 Ulrich Drepper's @file{elfutils} package, starting with version 0.53, contains
20226 a version of the @code{strip} command such that the command @kbd{strip foo -f
20227 foo.debug} has the same functionality as the two @code{objcopy} commands and
20228 the @code{ln -s} command above, together.
20231 Build ID gets embedded into the main executable using @code{ld --build-id} or
20232 the @value{NGCC} counterpart @code{gcc -Wl,--build-id}. Build ID support plus
20233 compatibility fixes for debug files separation are present in @sc{gnu} binary
20234 utilities (Binutils) package since version 2.18.
20239 @cindex CRC algorithm definition
20240 The CRC used in @code{.gnu_debuglink} is the CRC-32 defined in
20241 IEEE 802.3 using the polynomial:
20243 @c TexInfo requires naked braces for multi-digit exponents for Tex
20244 @c output, but this causes HTML output to barf. HTML has to be set using
20245 @c raw commands. So we end up having to specify this equation in 2
20250 <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>
20251 + <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
20257 @math{x^{32} + x^{26} + x^{23} + x^{22} + x^{16} + x^{12} + x^{11}}
20258 @math{+ x^{10} + x^8 + x^7 + x^5 + x^4 + x^2 + x + 1}
20262 The function is computed byte at a time, taking the least
20263 significant bit of each byte first. The initial pattern
20264 @code{0xffffffff} is used, to ensure leading zeros affect the CRC and
20265 the final result is inverted to ensure trailing zeros also affect the
20268 @emph{Note:} This is the same CRC polynomial as used in handling the
20269 @dfn{Remote Serial Protocol} @code{qCRC} packet (@pxref{qCRC packet}).
20270 However in the case of the Remote Serial Protocol, the CRC is computed
20271 @emph{most} significant bit first, and the result is not inverted, so
20272 trailing zeros have no effect on the CRC value.
20274 To complete the description, we show below the code of the function
20275 which produces the CRC used in @code{.gnu_debuglink}. Inverting the
20276 initially supplied @code{crc} argument means that an initial call to
20277 this function passing in zero will start computing the CRC using
20280 @kindex gnu_debuglink_crc32
20283 gnu_debuglink_crc32 (unsigned long crc,
20284 unsigned char *buf, size_t len)
20286 static const unsigned long crc32_table[256] =
20288 0x00000000, 0x77073096, 0xee0e612c, 0x990951ba, 0x076dc419,
20289 0x706af48f, 0xe963a535, 0x9e6495a3, 0x0edb8832, 0x79dcb8a4,
20290 0xe0d5e91e, 0x97d2d988, 0x09b64c2b, 0x7eb17cbd, 0xe7b82d07,
20291 0x90bf1d91, 0x1db71064, 0x6ab020f2, 0xf3b97148, 0x84be41de,
20292 0x1adad47d, 0x6ddde4eb, 0xf4d4b551, 0x83d385c7, 0x136c9856,
20293 0x646ba8c0, 0xfd62f97a, 0x8a65c9ec, 0x14015c4f, 0x63066cd9,
20294 0xfa0f3d63, 0x8d080df5, 0x3b6e20c8, 0x4c69105e, 0xd56041e4,
20295 0xa2677172, 0x3c03e4d1, 0x4b04d447, 0xd20d85fd, 0xa50ab56b,
20296 0x35b5a8fa, 0x42b2986c, 0xdbbbc9d6, 0xacbcf940, 0x32d86ce3,
20297 0x45df5c75, 0xdcd60dcf, 0xabd13d59, 0x26d930ac, 0x51de003a,
20298 0xc8d75180, 0xbfd06116, 0x21b4f4b5, 0x56b3c423, 0xcfba9599,
20299 0xb8bda50f, 0x2802b89e, 0x5f058808, 0xc60cd9b2, 0xb10be924,
20300 0x2f6f7c87, 0x58684c11, 0xc1611dab, 0xb6662d3d, 0x76dc4190,
20301 0x01db7106, 0x98d220bc, 0xefd5102a, 0x71b18589, 0x06b6b51f,
20302 0x9fbfe4a5, 0xe8b8d433, 0x7807c9a2, 0x0f00f934, 0x9609a88e,
20303 0xe10e9818, 0x7f6a0dbb, 0x086d3d2d, 0x91646c97, 0xe6635c01,
20304 0x6b6b51f4, 0x1c6c6162, 0x856530d8, 0xf262004e, 0x6c0695ed,
20305 0x1b01a57b, 0x8208f4c1, 0xf50fc457, 0x65b0d9c6, 0x12b7e950,
20306 0x8bbeb8ea, 0xfcb9887c, 0x62dd1ddf, 0x15da2d49, 0x8cd37cf3,
20307 0xfbd44c65, 0x4db26158, 0x3ab551ce, 0xa3bc0074, 0xd4bb30e2,
20308 0x4adfa541, 0x3dd895d7, 0xa4d1c46d, 0xd3d6f4fb, 0x4369e96a,
20309 0x346ed9fc, 0xad678846, 0xda60b8d0, 0x44042d73, 0x33031de5,
20310 0xaa0a4c5f, 0xdd0d7cc9, 0x5005713c, 0x270241aa, 0xbe0b1010,
20311 0xc90c2086, 0x5768b525, 0x206f85b3, 0xb966d409, 0xce61e49f,
20312 0x5edef90e, 0x29d9c998, 0xb0d09822, 0xc7d7a8b4, 0x59b33d17,
20313 0x2eb40d81, 0xb7bd5c3b, 0xc0ba6cad, 0xedb88320, 0x9abfb3b6,
20314 0x03b6e20c, 0x74b1d29a, 0xead54739, 0x9dd277af, 0x04db2615,
20315 0x73dc1683, 0xe3630b12, 0x94643b84, 0x0d6d6a3e, 0x7a6a5aa8,
20316 0xe40ecf0b, 0x9309ff9d, 0x0a00ae27, 0x7d079eb1, 0xf00f9344,
20317 0x8708a3d2, 0x1e01f268, 0x6906c2fe, 0xf762575d, 0x806567cb,
20318 0x196c3671, 0x6e6b06e7, 0xfed41b76, 0x89d32be0, 0x10da7a5a,
20319 0x67dd4acc, 0xf9b9df6f, 0x8ebeeff9, 0x17b7be43, 0x60b08ed5,
20320 0xd6d6a3e8, 0xa1d1937e, 0x38d8c2c4, 0x4fdff252, 0xd1bb67f1,
20321 0xa6bc5767, 0x3fb506dd, 0x48b2364b, 0xd80d2bda, 0xaf0a1b4c,
20322 0x36034af6, 0x41047a60, 0xdf60efc3, 0xa867df55, 0x316e8eef,
20323 0x4669be79, 0xcb61b38c, 0xbc66831a, 0x256fd2a0, 0x5268e236,
20324 0xcc0c7795, 0xbb0b4703, 0x220216b9, 0x5505262f, 0xc5ba3bbe,
20325 0xb2bd0b28, 0x2bb45a92, 0x5cb36a04, 0xc2d7ffa7, 0xb5d0cf31,
20326 0x2cd99e8b, 0x5bdeae1d, 0x9b64c2b0, 0xec63f226, 0x756aa39c,
20327 0x026d930a, 0x9c0906a9, 0xeb0e363f, 0x72076785, 0x05005713,
20328 0x95bf4a82, 0xe2b87a14, 0x7bb12bae, 0x0cb61b38, 0x92d28e9b,
20329 0xe5d5be0d, 0x7cdcefb7, 0x0bdbdf21, 0x86d3d2d4, 0xf1d4e242,
20330 0x68ddb3f8, 0x1fda836e, 0x81be16cd, 0xf6b9265b, 0x6fb077e1,
20331 0x18b74777, 0x88085ae6, 0xff0f6a70, 0x66063bca, 0x11010b5c,
20332 0x8f659eff, 0xf862ae69, 0x616bffd3, 0x166ccf45, 0xa00ae278,
20333 0xd70dd2ee, 0x4e048354, 0x3903b3c2, 0xa7672661, 0xd06016f7,
20334 0x4969474d, 0x3e6e77db, 0xaed16a4a, 0xd9d65adc, 0x40df0b66,
20335 0x37d83bf0, 0xa9bcae53, 0xdebb9ec5, 0x47b2cf7f, 0x30b5ffe9,
20336 0xbdbdf21c, 0xcabac28a, 0x53b39330, 0x24b4a3a6, 0xbad03605,
20337 0xcdd70693, 0x54de5729, 0x23d967bf, 0xb3667a2e, 0xc4614ab8,
20338 0x5d681b02, 0x2a6f2b94, 0xb40bbe37, 0xc30c8ea1, 0x5a05df1b,
20341 unsigned char *end;
20343 crc = ~crc & 0xffffffff;
20344 for (end = buf + len; buf < end; ++buf)
20345 crc = crc32_table[(crc ^ *buf) & 0xff] ^ (crc >> 8);
20346 return ~crc & 0xffffffff;
20351 This computation does not apply to the ``build ID'' method.
20353 @node MiniDebugInfo
20354 @section Debugging information in a special section
20355 @cindex separate debug sections
20356 @cindex @samp{.gnu_debugdata} section
20358 Some systems ship pre-built executables and libraries that have a
20359 special @samp{.gnu_debugdata} section. This feature is called
20360 @dfn{MiniDebugInfo}. This section holds an LZMA-compressed object and
20361 is used to supply extra symbols for backtraces.
20363 The intent of this section is to provide extra minimal debugging
20364 information for use in simple backtraces. It is not intended to be a
20365 replacement for full separate debugging information (@pxref{Separate
20366 Debug Files}). The example below shows the intended use; however,
20367 @value{GDBN} does not currently put restrictions on what sort of
20368 debugging information might be included in the section.
20370 @value{GDBN} has support for this extension. If the section exists,
20371 then it is used provided that no other source of debugging information
20372 can be found, and that @value{GDBN} was configured with LZMA support.
20374 This section can be easily created using @command{objcopy} and other
20375 standard utilities:
20378 # Extract the dynamic symbols from the main binary, there is no need
20379 # to also have these in the normal symbol table.
20380 nm -D @var{binary} --format=posix --defined-only \
20381 | awk '@{ print $1 @}' | sort > dynsyms
20383 # Extract all the text (i.e. function) symbols from the debuginfo.
20384 # (Note that we actually also accept "D" symbols, for the benefit
20385 # of platforms like PowerPC64 that use function descriptors.)
20386 nm @var{binary} --format=posix --defined-only \
20387 | awk '@{ if ($2 == "T" || $2 == "t" || $2 == "D") print $1 @}' \
20390 # Keep all the function symbols not already in the dynamic symbol
20392 comm -13 dynsyms funcsyms > keep_symbols
20394 # Separate full debug info into debug binary.
20395 objcopy --only-keep-debug @var{binary} debug
20397 # Copy the full debuginfo, keeping only a minimal set of symbols and
20398 # removing some unnecessary sections.
20399 objcopy -S --remove-section .gdb_index --remove-section .comment \
20400 --keep-symbols=keep_symbols debug mini_debuginfo
20402 # Drop the full debug info from the original binary.
20403 strip --strip-all -R .comment @var{binary}
20405 # Inject the compressed data into the .gnu_debugdata section of the
20408 objcopy --add-section .gnu_debugdata=mini_debuginfo.xz @var{binary}
20412 @section Index Files Speed Up @value{GDBN}
20413 @cindex index files
20414 @cindex @samp{.gdb_index} section
20416 When @value{GDBN} finds a symbol file, it scans the symbols in the
20417 file in order to construct an internal symbol table. This lets most
20418 @value{GDBN} operations work quickly---at the cost of a delay early
20419 on. For large programs, this delay can be quite lengthy, so
20420 @value{GDBN} provides a way to build an index, which speeds up
20423 For convenience, @value{GDBN} comes with a program,
20424 @command{gdb-add-index}, which can be used to add the index to a
20425 symbol file. It takes the symbol file as its only argument:
20428 $ gdb-add-index symfile
20431 @xref{gdb-add-index}.
20433 It is also possible to do the work manually. Here is what
20434 @command{gdb-add-index} does behind the curtains.
20436 The index is stored as a section in the symbol file. @value{GDBN} can
20437 write the index to a file, then you can put it into the symbol file
20438 using @command{objcopy}.
20440 To create an index file, use the @code{save gdb-index} command:
20443 @item save gdb-index [-dwarf-5] @var{directory}
20444 @kindex save gdb-index
20445 Create index files for all symbol files currently known by
20446 @value{GDBN}. For each known @var{symbol-file}, this command by
20447 default creates it produces a single file
20448 @file{@var{symbol-file}.gdb-index}. If you invoke this command with
20449 the @option{-dwarf-5} option, it produces 2 files:
20450 @file{@var{symbol-file}.debug_names} and
20451 @file{@var{symbol-file}.debug_str}. The files are created in the
20452 given @var{directory}.
20455 Once you have created an index file you can merge it into your symbol
20456 file, here named @file{symfile}, using @command{objcopy}:
20459 $ objcopy --add-section .gdb_index=symfile.gdb-index \
20460 --set-section-flags .gdb_index=readonly symfile symfile
20463 Or for @code{-dwarf-5}:
20466 $ objcopy --dump-section .debug_str=symfile.debug_str.new symfile
20467 $ cat symfile.debug_str >>symfile.debug_str.new
20468 $ objcopy --add-section .debug_names=symfile.gdb-index \
20469 --set-section-flags .debug_names=readonly \
20470 --update-section .debug_str=symfile.debug_str.new symfile symfile
20473 @value{GDBN} will normally ignore older versions of @file{.gdb_index}
20474 sections that have been deprecated. Usually they are deprecated because
20475 they are missing a new feature or have performance issues.
20476 To tell @value{GDBN} to use a deprecated index section anyway
20477 specify @code{set use-deprecated-index-sections on}.
20478 The default is @code{off}.
20479 This can speed up startup, but may result in some functionality being lost.
20480 @xref{Index Section Format}.
20482 @emph{Warning:} Setting @code{use-deprecated-index-sections} to @code{on}
20483 must be done before gdb reads the file. The following will not work:
20486 $ gdb -ex "set use-deprecated-index-sections on" <program>
20489 Instead you must do, for example,
20492 $ gdb -iex "set use-deprecated-index-sections on" <program>
20495 There are currently some limitation on indices. They only work when
20496 for DWARF debugging information, not stabs. And, they do not
20497 currently work for programs using Ada.
20499 @subsection Automatic symbol index cache
20501 @cindex automatic symbol index cache
20502 It is possible for @value{GDBN} to automatically save a copy of this index in a
20503 cache on disk and retrieve it from there when loading the same binary in the
20504 future. This feature can be turned on with @kbd{set index-cache on}. The
20505 following commands can be used to tweak the behavior of the index cache.
20509 @kindex set index-cache
20510 @item set index-cache on
20511 @itemx set index-cache off
20512 Enable or disable the use of the symbol index cache.
20514 @item set index-cache directory @var{directory}
20515 @kindex show index-cache
20516 @itemx show index-cache directory
20517 Set/show the directory where index files will be saved.
20519 The default value for this directory depends on the host platform. On
20520 most systems, the index is cached in the @file{gdb} subdirectory of
20521 the directory pointed to by the @env{XDG_CACHE_HOME} environment
20522 variable, if it is defined, else in the @file{.cache/gdb} subdirectory
20523 of your home directory. However, on some systems, the default may
20524 differ according to local convention.
20526 There is no limit on the disk space used by index cache. It is perfectly safe
20527 to delete the content of that directory to free up disk space.
20529 @item show index-cache stats
20530 Print the number of cache hits and misses since the launch of @value{GDBN}.
20534 @node Symbol Errors
20535 @section Errors Reading Symbol Files
20537 While reading a symbol file, @value{GDBN} occasionally encounters problems,
20538 such as symbol types it does not recognize, or known bugs in compiler
20539 output. By default, @value{GDBN} does not notify you of such problems, since
20540 they are relatively common and primarily of interest to people
20541 debugging compilers. If you are interested in seeing information
20542 about ill-constructed symbol tables, you can either ask @value{GDBN} to print
20543 only one message about each such type of problem, no matter how many
20544 times the problem occurs; or you can ask @value{GDBN} to print more messages,
20545 to see how many times the problems occur, with the @code{set
20546 complaints} command (@pxref{Messages/Warnings, ,Optional Warnings and
20549 The messages currently printed, and their meanings, include:
20552 @item inner block not inside outer block in @var{symbol}
20554 The symbol information shows where symbol scopes begin and end
20555 (such as at the start of a function or a block of statements). This
20556 error indicates that an inner scope block is not fully contained
20557 in its outer scope blocks.
20559 @value{GDBN} circumvents the problem by treating the inner block as if it had
20560 the same scope as the outer block. In the error message, @var{symbol}
20561 may be shown as ``@code{(don't know)}'' if the outer block is not a
20564 @item block at @var{address} out of order
20566 The symbol information for symbol scope blocks should occur in
20567 order of increasing addresses. This error indicates that it does not
20570 @value{GDBN} does not circumvent this problem, and has trouble
20571 locating symbols in the source file whose symbols it is reading. (You
20572 can often determine what source file is affected by specifying
20573 @code{set verbose on}. @xref{Messages/Warnings, ,Optional Warnings and
20576 @item bad block start address patched
20578 The symbol information for a symbol scope block has a start address
20579 smaller than the address of the preceding source line. This is known
20580 to occur in the SunOS 4.1.1 (and earlier) C compiler.
20582 @value{GDBN} circumvents the problem by treating the symbol scope block as
20583 starting on the previous source line.
20585 @item bad string table offset in symbol @var{n}
20588 Symbol number @var{n} contains a pointer into the string table which is
20589 larger than the size of the string table.
20591 @value{GDBN} circumvents the problem by considering the symbol to have the
20592 name @code{foo}, which may cause other problems if many symbols end up
20595 @item unknown symbol type @code{0x@var{nn}}
20597 The symbol information contains new data types that @value{GDBN} does
20598 not yet know how to read. @code{0x@var{nn}} is the symbol type of the
20599 uncomprehended information, in hexadecimal.
20601 @value{GDBN} circumvents the error by ignoring this symbol information.
20602 This usually allows you to debug your program, though certain symbols
20603 are not accessible. If you encounter such a problem and feel like
20604 debugging it, you can debug @code{@value{GDBP}} with itself, breakpoint
20605 on @code{complain}, then go up to the function @code{read_dbx_symtab}
20606 and examine @code{*bufp} to see the symbol.
20608 @item stub type has NULL name
20610 @value{GDBN} could not find the full definition for a struct or class.
20612 @item const/volatile indicator missing (ok if using g++ v1.x), got@dots{}
20613 The symbol information for a C@t{++} member function is missing some
20614 information that recent versions of the compiler should have output for
20617 @item info mismatch between compiler and debugger
20619 @value{GDBN} could not parse a type specification output by the compiler.
20624 @section GDB Data Files
20626 @cindex prefix for data files
20627 @value{GDBN} will sometimes read an auxiliary data file. These files
20628 are kept in a directory known as the @dfn{data directory}.
20630 You can set the data directory's name, and view the name @value{GDBN}
20631 is currently using.
20634 @kindex set data-directory
20635 @item set data-directory @var{directory}
20636 Set the directory which @value{GDBN} searches for auxiliary data files
20637 to @var{directory}.
20639 @kindex show data-directory
20640 @item show data-directory
20641 Show the directory @value{GDBN} searches for auxiliary data files.
20644 @cindex default data directory
20645 @cindex @samp{--with-gdb-datadir}
20646 You can set the default data directory by using the configure-time
20647 @samp{--with-gdb-datadir} option. If the data directory is inside
20648 @value{GDBN}'s configured binary prefix (set with @samp{--prefix} or
20649 @samp{--exec-prefix}), then the default data directory will be updated
20650 automatically if the installed @value{GDBN} is moved to a new
20653 The data directory may also be specified with the
20654 @code{--data-directory} command line option.
20655 @xref{Mode Options}.
20658 @chapter Specifying a Debugging Target
20660 @cindex debugging target
20661 A @dfn{target} is the execution environment occupied by your program.
20663 Often, @value{GDBN} runs in the same host environment as your program;
20664 in that case, the debugging target is specified as a side effect when
20665 you use the @code{file} or @code{core} commands. When you need more
20666 flexibility---for example, running @value{GDBN} on a physically separate
20667 host, or controlling a standalone system over a serial port or a
20668 realtime system over a TCP/IP connection---you can use the @code{target}
20669 command to specify one of the target types configured for @value{GDBN}
20670 (@pxref{Target Commands, ,Commands for Managing Targets}).
20672 @cindex target architecture
20673 It is possible to build @value{GDBN} for several different @dfn{target
20674 architectures}. When @value{GDBN} is built like that, you can choose
20675 one of the available architectures with the @kbd{set architecture}
20679 @kindex set architecture
20680 @kindex show architecture
20681 @item set architecture @var{arch}
20682 This command sets the current target architecture to @var{arch}. The
20683 value of @var{arch} can be @code{"auto"}, in addition to one of the
20684 supported architectures.
20686 @item show architecture
20687 Show the current target architecture.
20689 @item set processor
20691 @kindex set processor
20692 @kindex show processor
20693 These are alias commands for, respectively, @code{set architecture}
20694 and @code{show architecture}.
20698 * Active Targets:: Active targets
20699 * Target Commands:: Commands for managing targets
20700 * Byte Order:: Choosing target byte order
20703 @node Active Targets
20704 @section Active Targets
20706 @cindex stacking targets
20707 @cindex active targets
20708 @cindex multiple targets
20710 There are multiple classes of targets such as: processes, executable files or
20711 recording sessions. Core files belong to the process class, making core file
20712 and process mutually exclusive. Otherwise, @value{GDBN} can work concurrently
20713 on multiple active targets, one in each class. This allows you to (for
20714 example) start a process and inspect its activity, while still having access to
20715 the executable file after the process finishes. Or if you start process
20716 recording (@pxref{Reverse Execution}) and @code{reverse-step} there, you are
20717 presented a virtual layer of the recording target, while the process target
20718 remains stopped at the chronologically last point of the process execution.
20720 Use the @code{core-file} and @code{exec-file} commands to select a new core
20721 file or executable target (@pxref{Files, ,Commands to Specify Files}). To
20722 specify as a target a process that is already running, use the @code{attach}
20723 command (@pxref{Attach, ,Debugging an Already-running Process}).
20725 @node Target Commands
20726 @section Commands for Managing Targets
20729 @item target @var{type} @var{parameters}
20730 Connects the @value{GDBN} host environment to a target machine or
20731 process. A target is typically a protocol for talking to debugging
20732 facilities. You use the argument @var{type} to specify the type or
20733 protocol of the target machine.
20735 Further @var{parameters} are interpreted by the target protocol, but
20736 typically include things like device names or host names to connect
20737 with, process numbers, and baud rates.
20739 The @code{target} command does not repeat if you press @key{RET} again
20740 after executing the command.
20742 @kindex help target
20744 Displays the names of all targets available. To display targets
20745 currently selected, use either @code{info target} or @code{info files}
20746 (@pxref{Files, ,Commands to Specify Files}).
20748 @item help target @var{name}
20749 Describe a particular target, including any parameters necessary to
20752 @kindex set gnutarget
20753 @item set gnutarget @var{args}
20754 @value{GDBN} uses its own library BFD to read your files. @value{GDBN}
20755 knows whether it is reading an @dfn{executable},
20756 a @dfn{core}, or a @dfn{.o} file; however, you can specify the file format
20757 with the @code{set gnutarget} command. Unlike most @code{target} commands,
20758 with @code{gnutarget} the @code{target} refers to a program, not a machine.
20761 @emph{Warning:} To specify a file format with @code{set gnutarget},
20762 you must know the actual BFD name.
20766 @xref{Files, , Commands to Specify Files}.
20768 @kindex show gnutarget
20769 @item show gnutarget
20770 Use the @code{show gnutarget} command to display what file format
20771 @code{gnutarget} is set to read. If you have not set @code{gnutarget},
20772 @value{GDBN} will determine the file format for each file automatically,
20773 and @code{show gnutarget} displays @samp{The current BFD target is "auto"}.
20776 @cindex common targets
20777 Here are some common targets (available, or not, depending on the GDB
20782 @item target exec @var{program}
20783 @cindex executable file target
20784 An executable file. @samp{target exec @var{program}} is the same as
20785 @samp{exec-file @var{program}}.
20787 @item target core @var{filename}
20788 @cindex core dump file target
20789 A core dump file. @samp{target core @var{filename}} is the same as
20790 @samp{core-file @var{filename}}.
20792 @item target remote @var{medium}
20793 @cindex remote target
20794 A remote system connected to @value{GDBN} via a serial line or network
20795 connection. This command tells @value{GDBN} to use its own remote
20796 protocol over @var{medium} for debugging. @xref{Remote Debugging}.
20798 For example, if you have a board connected to @file{/dev/ttya} on the
20799 machine running @value{GDBN}, you could say:
20802 target remote /dev/ttya
20805 @code{target remote} supports the @code{load} command. This is only
20806 useful if you have some other way of getting the stub to the target
20807 system, and you can put it somewhere in memory where it won't get
20808 clobbered by the download.
20810 @item target sim @r{[}@var{simargs}@r{]} @dots{}
20811 @cindex built-in simulator target
20812 Builtin CPU simulator. @value{GDBN} includes simulators for most architectures.
20820 works; however, you cannot assume that a specific memory map, device
20821 drivers, or even basic I/O is available, although some simulators do
20822 provide these. For info about any processor-specific simulator details,
20823 see the appropriate section in @ref{Embedded Processors, ,Embedded
20826 @item target native
20827 @cindex native target
20828 Setup for local/native process debugging. Useful to make the
20829 @code{run} command spawn native processes (likewise @code{attach},
20830 etc.@:) even when @code{set auto-connect-native-target} is @code{off}
20831 (@pxref{set auto-connect-native-target}).
20835 Different targets are available on different configurations of @value{GDBN};
20836 your configuration may have more or fewer targets.
20838 Many remote targets require you to download the executable's code once
20839 you've successfully established a connection. You may wish to control
20840 various aspects of this process.
20845 @kindex set hash@r{, for remote monitors}
20846 @cindex hash mark while downloading
20847 This command controls whether a hash mark @samp{#} is displayed while
20848 downloading a file to the remote monitor. If on, a hash mark is
20849 displayed after each S-record is successfully downloaded to the
20853 @kindex show hash@r{, for remote monitors}
20854 Show the current status of displaying the hash mark.
20856 @item set debug monitor
20857 @kindex set debug monitor
20858 @cindex display remote monitor communications
20859 Enable or disable display of communications messages between
20860 @value{GDBN} and the remote monitor.
20862 @item show debug monitor
20863 @kindex show debug monitor
20864 Show the current status of displaying communications between
20865 @value{GDBN} and the remote monitor.
20870 @kindex load @var{filename} @var{offset}
20871 @item load @var{filename} @var{offset}
20873 Depending on what remote debugging facilities are configured into
20874 @value{GDBN}, the @code{load} command may be available. Where it exists, it
20875 is meant to make @var{filename} (an executable) available for debugging
20876 on the remote system---by downloading, or dynamic linking, for example.
20877 @code{load} also records the @var{filename} symbol table in @value{GDBN}, like
20878 the @code{add-symbol-file} command.
20880 If your @value{GDBN} does not have a @code{load} command, attempting to
20881 execute it gets the error message ``@code{You can't do that when your
20882 target is @dots{}}''
20884 The file is loaded at whatever address is specified in the executable.
20885 For some object file formats, you can specify the load address when you
20886 link the program; for other formats, like a.out, the object file format
20887 specifies a fixed address.
20888 @c FIXME! This would be a good place for an xref to the GNU linker doc.
20890 It is also possible to tell @value{GDBN} to load the executable file at a
20891 specific offset described by the optional argument @var{offset}. When
20892 @var{offset} is provided, @var{filename} must also be provided.
20894 Depending on the remote side capabilities, @value{GDBN} may be able to
20895 load programs into flash memory.
20897 @code{load} does not repeat if you press @key{RET} again after using it.
20902 @kindex flash-erase
20904 @anchor{flash-erase}
20906 Erases all known flash memory regions on the target.
20911 @section Choosing Target Byte Order
20913 @cindex choosing target byte order
20914 @cindex target byte order
20916 Some types of processors, such as the @acronym{MIPS}, PowerPC, and Renesas SH,
20917 offer the ability to run either big-endian or little-endian byte
20918 orders. Usually the executable or symbol will include a bit to
20919 designate the endian-ness, and you will not need to worry about
20920 which to use. However, you may still find it useful to adjust
20921 @value{GDBN}'s idea of processor endian-ness manually.
20925 @item set endian big
20926 Instruct @value{GDBN} to assume the target is big-endian.
20928 @item set endian little
20929 Instruct @value{GDBN} to assume the target is little-endian.
20931 @item set endian auto
20932 Instruct @value{GDBN} to use the byte order associated with the
20936 Display @value{GDBN}'s current idea of the target byte order.
20940 If the @code{set endian auto} mode is in effect and no executable has
20941 been selected, then the endianness used is the last one chosen either
20942 by one of the @code{set endian big} and @code{set endian little}
20943 commands or by inferring from the last executable used. If no
20944 endianness has been previously chosen, then the default for this mode
20945 is inferred from the target @value{GDBN} has been built for, and is
20946 @code{little} if the name of the target CPU has an @code{el} suffix
20947 and @code{big} otherwise.
20949 Note that these commands merely adjust interpretation of symbolic
20950 data on the host, and that they have absolutely no effect on the
20954 @node Remote Debugging
20955 @chapter Debugging Remote Programs
20956 @cindex remote debugging
20958 If you are trying to debug a program running on a machine that cannot run
20959 @value{GDBN} in the usual way, it is often useful to use remote debugging.
20960 For example, you might use remote debugging on an operating system kernel,
20961 or on a small system which does not have a general purpose operating system
20962 powerful enough to run a full-featured debugger.
20964 Some configurations of @value{GDBN} have special serial or TCP/IP interfaces
20965 to make this work with particular debugging targets. In addition,
20966 @value{GDBN} comes with a generic serial protocol (specific to @value{GDBN},
20967 but not specific to any particular target system) which you can use if you
20968 write the remote stubs---the code that runs on the remote system to
20969 communicate with @value{GDBN}.
20971 Other remote targets may be available in your
20972 configuration of @value{GDBN}; use @code{help target} to list them.
20975 * Connecting:: Connecting to a remote target
20976 * File Transfer:: Sending files to a remote system
20977 * Server:: Using the gdbserver program
20978 * Remote Configuration:: Remote configuration
20979 * Remote Stub:: Implementing a remote stub
20983 @section Connecting to a Remote Target
20984 @cindex remote debugging, connecting
20985 @cindex @code{gdbserver}, connecting
20986 @cindex remote debugging, types of connections
20987 @cindex @code{gdbserver}, types of connections
20988 @cindex @code{gdbserver}, @code{target remote} mode
20989 @cindex @code{gdbserver}, @code{target extended-remote} mode
20991 This section describes how to connect to a remote target, including the
20992 types of connections and their differences, how to set up executable and
20993 symbol files on the host and target, and the commands used for
20994 connecting to and disconnecting from the remote target.
20996 @subsection Types of Remote Connections
20998 @value{GDBN} supports two types of remote connections, @code{target remote}
20999 mode and @code{target extended-remote} mode. Note that many remote targets
21000 support only @code{target remote} mode. There are several major
21001 differences between the two types of connections, enumerated here:
21005 @cindex remote debugging, detach and program exit
21006 @item Result of detach or program exit
21007 @strong{With target remote mode:} When the debugged program exits or you
21008 detach from it, @value{GDBN} disconnects from the target. When using
21009 @code{gdbserver}, @code{gdbserver} will exit.
21011 @strong{With target extended-remote mode:} When the debugged program exits or
21012 you detach from it, @value{GDBN} remains connected to the target, even
21013 though no program is running. You can rerun the program, attach to a
21014 running program, or use @code{monitor} commands specific to the target.
21016 When using @code{gdbserver} in this case, it does not exit unless it was
21017 invoked using the @option{--once} option. If the @option{--once} option
21018 was not used, you can ask @code{gdbserver} to exit using the
21019 @code{monitor exit} command (@pxref{Monitor Commands for gdbserver}).
21021 @item Specifying the program to debug
21022 For both connection types you use the @code{file} command to specify the
21023 program on the host system. If you are using @code{gdbserver} there are
21024 some differences in how to specify the location of the program on the
21027 @strong{With target remote mode:} You must either specify the program to debug
21028 on the @code{gdbserver} command line or use the @option{--attach} option
21029 (@pxref{Attaching to a program,,Attaching to a Running Program}).
21031 @cindex @option{--multi}, @code{gdbserver} option
21032 @strong{With target extended-remote mode:} You may specify the program to debug
21033 on the @code{gdbserver} command line, or you can load the program or attach
21034 to it using @value{GDBN} commands after connecting to @code{gdbserver}.
21036 @anchor{--multi Option in Types of Remote Connnections}
21037 You can start @code{gdbserver} without supplying an initial command to run
21038 or process ID to attach. To do this, use the @option{--multi} command line
21039 option. Then you can connect using @code{target extended-remote} and start
21040 the program you want to debug (see below for details on using the
21041 @code{run} command in this scenario). Note that the conditions under which
21042 @code{gdbserver} terminates depend on how @value{GDBN} connects to it
21043 (@code{target remote} or @code{target extended-remote}). The
21044 @option{--multi} option to @code{gdbserver} has no influence on that.
21046 @item The @code{run} command
21047 @strong{With target remote mode:} The @code{run} command is not
21048 supported. Once a connection has been established, you can use all
21049 the usual @value{GDBN} commands to examine and change data. The
21050 remote program is already running, so you can use commands like
21051 @kbd{step} and @kbd{continue}.
21053 @strong{With target extended-remote mode:} The @code{run} command is
21054 supported. The @code{run} command uses the value set by
21055 @code{set remote exec-file} (@pxref{set remote exec-file}) to select
21056 the program to run. Command line arguments are supported, except for
21057 wildcard expansion and I/O redirection (@pxref{Arguments}).
21059 If you specify the program to debug on the command line, then the
21060 @code{run} command is not required to start execution, and you can
21061 resume using commands like @kbd{step} and @kbd{continue} as with
21062 @code{target remote} mode.
21064 @anchor{Attaching in Types of Remote Connections}
21066 @strong{With target remote mode:} The @value{GDBN} command @code{attach} is
21067 not supported. To attach to a running program using @code{gdbserver}, you
21068 must use the @option{--attach} option (@pxref{Running gdbserver}).
21070 @strong{With target extended-remote mode:} To attach to a running program,
21071 you may use the @code{attach} command after the connection has been
21072 established. If you are using @code{gdbserver}, you may also invoke
21073 @code{gdbserver} using the @option{--attach} option
21074 (@pxref{Running gdbserver}).
21078 @anchor{Host and target files}
21079 @subsection Host and Target Files
21080 @cindex remote debugging, symbol files
21081 @cindex symbol files, remote debugging
21083 @value{GDBN}, running on the host, needs access to symbol and debugging
21084 information for your program running on the target. This requires
21085 access to an unstripped copy of your program, and possibly any associated
21086 symbol files. Note that this section applies equally to both @code{target
21087 remote} mode and @code{target extended-remote} mode.
21089 Some remote targets (@pxref{qXfer executable filename read}, and
21090 @pxref{Host I/O Packets}) allow @value{GDBN} to access program files over
21091 the same connection used to communicate with @value{GDBN}. With such a
21092 target, if the remote program is unstripped, the only command you need is
21093 @code{target remote} (or @code{target extended-remote}).
21095 If the remote program is stripped, or the target does not support remote
21096 program file access, start up @value{GDBN} using the name of the local
21097 unstripped copy of your program as the first argument, or use the
21098 @code{file} command. Use @code{set sysroot} to specify the location (on
21099 the host) of target libraries (unless your @value{GDBN} was compiled with
21100 the correct sysroot using @code{--with-sysroot}). Alternatively, you
21101 may use @code{set solib-search-path} to specify how @value{GDBN} locates
21104 The symbol file and target libraries must exactly match the executable
21105 and libraries on the target, with one exception: the files on the host
21106 system should not be stripped, even if the files on the target system
21107 are. Mismatched or missing files will lead to confusing results
21108 during debugging. On @sc{gnu}/Linux targets, mismatched or missing
21109 files may also prevent @code{gdbserver} from debugging multi-threaded
21112 @subsection Remote Connection Commands
21113 @cindex remote connection commands
21114 @value{GDBN} can communicate with the target over a serial line, a
21115 local Unix domain socket, or
21116 over an @acronym{IP} network using @acronym{TCP} or @acronym{UDP}. In
21117 each case, @value{GDBN} uses the same protocol for debugging your
21118 program; only the medium carrying the debugging packets varies. The
21119 @code{target remote} and @code{target extended-remote} commands
21120 establish a connection to the target. Both commands accept the same
21121 arguments, which indicate the medium to use:
21125 @item target remote @var{serial-device}
21126 @itemx target extended-remote @var{serial-device}
21127 @cindex serial line, @code{target remote}
21128 Use @var{serial-device} to communicate with the target. For example,
21129 to use a serial line connected to the device named @file{/dev/ttyb}:
21132 target remote /dev/ttyb
21135 If you're using a serial line, you may want to give @value{GDBN} the
21136 @samp{--baud} option, or use the @code{set serial baud} command
21137 (@pxref{Remote Configuration, set serial baud}) before the
21138 @code{target} command.
21140 @item target remote @var{local-socket}
21141 @itemx target extended-remote @var{local-socket}
21142 @cindex local socket, @code{target remote}
21143 @cindex Unix domain socket
21144 Use @var{local-socket} to communicate with the target. For example,
21145 to use a local Unix domain socket bound to the file system entry @file{/tmp/gdb-socket0}:
21148 target remote /tmp/gdb-socket0
21151 Note that this command has the same form as the command to connect
21152 to a serial line. @value{GDBN} will automatically determine which
21153 kind of file you have specified and will make the appropriate kind
21155 This feature is not available if the host system does not support
21156 Unix domain sockets.
21158 @item target remote @code{@var{host}:@var{port}}
21159 @itemx target remote @code{@var{[host]}:@var{port}}
21160 @itemx target remote @code{tcp:@var{host}:@var{port}}
21161 @itemx target remote @code{tcp:@var{[host]}:@var{port}}
21162 @itemx target remote @code{tcp4:@var{host}:@var{port}}
21163 @itemx target remote @code{tcp6:@var{host}:@var{port}}
21164 @itemx target remote @code{tcp6:@var{[host]}:@var{port}}
21165 @itemx target extended-remote @code{@var{host}:@var{port}}
21166 @itemx target extended-remote @code{@var{[host]}:@var{port}}
21167 @itemx target extended-remote @code{tcp:@var{host}:@var{port}}
21168 @itemx target extended-remote @code{tcp:@var{[host]}:@var{port}}
21169 @itemx target extended-remote @code{tcp4:@var{host}:@var{port}}
21170 @itemx target extended-remote @code{tcp6:@var{host}:@var{port}}
21171 @itemx target extended-remote @code{tcp6:@var{[host]}:@var{port}}
21172 @cindex @acronym{TCP} port, @code{target remote}
21173 Debug using a @acronym{TCP} connection to @var{port} on @var{host}.
21174 The @var{host} may be either a host name, a numeric @acronym{IPv4}
21175 address, or a numeric @acronym{IPv6} address (with or without the
21176 square brackets to separate the address from the port); @var{port}
21177 must be a decimal number. The @var{host} could be the target machine
21178 itself, if it is directly connected to the net, or it might be a
21179 terminal server which in turn has a serial line to the target.
21181 For example, to connect to port 2828 on a terminal server named
21185 target remote manyfarms:2828
21188 To connect to port 2828 on a terminal server whose address is
21189 @code{2001:0db8:85a3:0000:0000:8a2e:0370:7334}, you can either use the
21190 square bracket syntax:
21193 target remote [2001:0db8:85a3:0000:0000:8a2e:0370:7334]:2828
21197 or explicitly specify the @acronym{IPv6} protocol:
21200 target remote tcp6:2001:0db8:85a3:0000:0000:8a2e:0370:7334:2828
21203 This last example may be confusing to the reader, because there is no
21204 visible separation between the hostname and the port number.
21205 Therefore, we recommend the user to provide @acronym{IPv6} addresses
21206 using square brackets for clarity. However, it is important to
21207 mention that for @value{GDBN} there is no ambiguity: the number after
21208 the last colon is considered to be the port number.
21210 If your remote target is actually running on the same machine as your
21211 debugger session (e.g.@: a simulator for your target running on the
21212 same host), you can omit the hostname. For example, to connect to
21213 port 1234 on your local machine:
21216 target remote :1234
21220 Note that the colon is still required here.
21222 @item target remote @code{udp:@var{host}:@var{port}}
21223 @itemx target remote @code{udp:@var{[host]}:@var{port}}
21224 @itemx target remote @code{udp4:@var{host}:@var{port}}
21225 @itemx target remote @code{udp6:@var{[host]}:@var{port}}
21226 @itemx target extended-remote @code{udp:@var{host}:@var{port}}
21227 @itemx target extended-remote @code{udp:@var{host}:@var{port}}
21228 @itemx target extended-remote @code{udp:@var{[host]}:@var{port}}
21229 @itemx target extended-remote @code{udp4:@var{host}:@var{port}}
21230 @itemx target extended-remote @code{udp6:@var{host}:@var{port}}
21231 @itemx target extended-remote @code{udp6:@var{[host]}:@var{port}}
21232 @cindex @acronym{UDP} port, @code{target remote}
21233 Debug using @acronym{UDP} packets to @var{port} on @var{host}. For example, to
21234 connect to @acronym{UDP} port 2828 on a terminal server named @code{manyfarms}:
21237 target remote udp:manyfarms:2828
21240 When using a @acronym{UDP} connection for remote debugging, you should
21241 keep in mind that the `U' stands for ``Unreliable''. @acronym{UDP}
21242 can silently drop packets on busy or unreliable networks, which will
21243 cause havoc with your debugging session.
21245 @item target remote | @var{command}
21246 @itemx target extended-remote | @var{command}
21247 @cindex pipe, @code{target remote} to
21248 Run @var{command} in the background and communicate with it using a
21249 pipe. The @var{command} is a shell command, to be parsed and expanded
21250 by the system's command shell, @code{/bin/sh}; it should expect remote
21251 protocol packets on its standard input, and send replies on its
21252 standard output. You could use this to run a stand-alone simulator
21253 that speaks the remote debugging protocol, to make net connections
21254 using programs like @code{ssh}, or for other similar tricks.
21256 If @var{command} closes its standard output (perhaps by exiting),
21257 @value{GDBN} will try to send it a @code{SIGTERM} signal. (If the
21258 program has already exited, this will have no effect.)
21262 @cindex interrupting remote programs
21263 @cindex remote programs, interrupting
21264 Whenever @value{GDBN} is waiting for the remote program, if you type the
21265 interrupt character (often @kbd{Ctrl-c}), @value{GDBN} attempts to stop the
21266 program. This may or may not succeed, depending in part on the hardware
21267 and the serial drivers the remote system uses. If you type the
21268 interrupt character once again, @value{GDBN} displays this prompt:
21271 Interrupted while waiting for the program.
21272 Give up (and stop debugging it)? (y or n)
21275 In @code{target remote} mode, if you type @kbd{y}, @value{GDBN} abandons
21276 the remote debugging session. (If you decide you want to try again later,
21277 you can use @kbd{target remote} again to connect once more.) If you type
21278 @kbd{n}, @value{GDBN} goes back to waiting.
21280 In @code{target extended-remote} mode, typing @kbd{n} will leave
21281 @value{GDBN} connected to the target.
21284 @kindex detach (remote)
21286 When you have finished debugging the remote program, you can use the
21287 @code{detach} command to release it from @value{GDBN} control.
21288 Detaching from the target normally resumes its execution, but the results
21289 will depend on your particular remote stub. After the @code{detach}
21290 command in @code{target remote} mode, @value{GDBN} is free to connect to
21291 another target. In @code{target extended-remote} mode, @value{GDBN} is
21292 still connected to the target.
21296 The @code{disconnect} command closes the connection to the target, and
21297 the target is generally not resumed. It will wait for @value{GDBN}
21298 (this instance or another one) to connect and continue debugging. After
21299 the @code{disconnect} command, @value{GDBN} is again free to connect to
21302 @cindex send command to remote monitor
21303 @cindex extend @value{GDBN} for remote targets
21304 @cindex add new commands for external monitor
21306 @item monitor @var{cmd}
21307 This command allows you to send arbitrary commands directly to the
21308 remote monitor. Since @value{GDBN} doesn't care about the commands it
21309 sends like this, this command is the way to extend @value{GDBN}---you
21310 can add new commands that only the external monitor will understand
21314 @node File Transfer
21315 @section Sending files to a remote system
21316 @cindex remote target, file transfer
21317 @cindex file transfer
21318 @cindex sending files to remote systems
21320 Some remote targets offer the ability to transfer files over the same
21321 connection used to communicate with @value{GDBN}. This is convenient
21322 for targets accessible through other means, e.g.@: @sc{gnu}/Linux systems
21323 running @code{gdbserver} over a network interface. For other targets,
21324 e.g.@: embedded devices with only a single serial port, this may be
21325 the only way to upload or download files.
21327 Not all remote targets support these commands.
21331 @item remote put @var{hostfile} @var{targetfile}
21332 Copy file @var{hostfile} from the host system (the machine running
21333 @value{GDBN}) to @var{targetfile} on the target system.
21336 @item remote get @var{targetfile} @var{hostfile}
21337 Copy file @var{targetfile} from the target system to @var{hostfile}
21338 on the host system.
21340 @kindex remote delete
21341 @item remote delete @var{targetfile}
21342 Delete @var{targetfile} from the target system.
21347 @section Using the @code{gdbserver} Program
21350 @cindex remote connection without stubs
21351 @code{gdbserver} is a control program for Unix-like systems, which
21352 allows you to connect your program with a remote @value{GDBN} via
21353 @code{target remote} or @code{target extended-remote}---but without
21354 linking in the usual debugging stub.
21356 @code{gdbserver} is not a complete replacement for the debugging stubs,
21357 because it requires essentially the same operating-system facilities
21358 that @value{GDBN} itself does. In fact, a system that can run
21359 @code{gdbserver} to connect to a remote @value{GDBN} could also run
21360 @value{GDBN} locally! @code{gdbserver} is sometimes useful nevertheless,
21361 because it is a much smaller program than @value{GDBN} itself. It is
21362 also easier to port than all of @value{GDBN}, so you may be able to get
21363 started more quickly on a new system by using @code{gdbserver}.
21364 Finally, if you develop code for real-time systems, you may find that
21365 the tradeoffs involved in real-time operation make it more convenient to
21366 do as much development work as possible on another system, for example
21367 by cross-compiling. You can use @code{gdbserver} to make a similar
21368 choice for debugging.
21370 @value{GDBN} and @code{gdbserver} communicate via either a serial line
21371 or a TCP connection, using the standard @value{GDBN} remote serial
21375 @emph{Warning:} @code{gdbserver} does not have any built-in security.
21376 Do not run @code{gdbserver} connected to any public network; a
21377 @value{GDBN} connection to @code{gdbserver} provides access to the
21378 target system with the same privileges as the user running
21382 @anchor{Running gdbserver}
21383 @subsection Running @code{gdbserver}
21384 @cindex arguments, to @code{gdbserver}
21385 @cindex @code{gdbserver}, command-line arguments
21387 Run @code{gdbserver} on the target system. You need a copy of the
21388 program you want to debug, including any libraries it requires.
21389 @code{gdbserver} does not need your program's symbol table, so you can
21390 strip the program if necessary to save space. @value{GDBN} on the host
21391 system does all the symbol handling.
21393 To use the server, you must tell it how to communicate with @value{GDBN};
21394 the name of your program; and the arguments for your program. The usual
21398 target> gdbserver @var{comm} @var{program} [ @var{args} @dots{} ]
21401 @var{comm} is either a device name (to use a serial line), or a TCP
21402 hostname and portnumber, or @code{-} or @code{stdio} to use
21403 stdin/stdout of @code{gdbserver}.
21404 For example, to debug Emacs with the argument
21405 @samp{foo.txt} and communicate with @value{GDBN} over the serial port
21409 target> gdbserver /dev/com1 emacs foo.txt
21412 @code{gdbserver} waits passively for the host @value{GDBN} to communicate
21415 To use a TCP connection instead of a serial line:
21418 target> gdbserver host:2345 emacs foo.txt
21421 The only difference from the previous example is the first argument,
21422 specifying that you are communicating with the host @value{GDBN} via
21423 TCP. The @samp{host:2345} argument means that @code{gdbserver} is to
21424 expect a TCP connection from machine @samp{host} to local TCP port 2345.
21425 (Currently, the @samp{host} part is ignored.) You can choose any number
21426 you want for the port number as long as it does not conflict with any
21427 TCP ports already in use on the target system (for example, @code{23} is
21428 reserved for @code{telnet}).@footnote{If you choose a port number that
21429 conflicts with another service, @code{gdbserver} prints an error message
21430 and exits.} You must use the same port number with the host @value{GDBN}
21431 @code{target remote} command.
21433 The @code{stdio} connection is useful when starting @code{gdbserver}
21437 (gdb) target remote | ssh -T hostname gdbserver - hello
21440 The @samp{-T} option to ssh is provided because we don't need a remote pty,
21441 and we don't want escape-character handling. Ssh does this by default when
21442 a command is provided, the flag is provided to make it explicit.
21443 You could elide it if you want to.
21445 Programs started with stdio-connected gdbserver have @file{/dev/null} for
21446 @code{stdin}, and @code{stdout},@code{stderr} are sent back to gdb for
21447 display through a pipe connected to gdbserver.
21448 Both @code{stdout} and @code{stderr} use the same pipe.
21450 @anchor{Attaching to a program}
21451 @subsubsection Attaching to a Running Program
21452 @cindex attach to a program, @code{gdbserver}
21453 @cindex @option{--attach}, @code{gdbserver} option
21455 On some targets, @code{gdbserver} can also attach to running programs.
21456 This is accomplished via the @code{--attach} argument. The syntax is:
21459 target> gdbserver --attach @var{comm} @var{pid}
21462 @var{pid} is the process ID of a currently running process. It isn't
21463 necessary to point @code{gdbserver} at a binary for the running process.
21465 In @code{target extended-remote} mode, you can also attach using the
21466 @value{GDBN} attach command
21467 (@pxref{Attaching in Types of Remote Connections}).
21470 You can debug processes by name instead of process ID if your target has the
21471 @code{pidof} utility:
21474 target> gdbserver --attach @var{comm} `pidof @var{program}`
21477 In case more than one copy of @var{program} is running, or @var{program}
21478 has multiple threads, most versions of @code{pidof} support the
21479 @code{-s} option to only return the first process ID.
21481 @subsubsection TCP port allocation lifecycle of @code{gdbserver}
21483 This section applies only when @code{gdbserver} is run to listen on a TCP
21486 @code{gdbserver} normally terminates after all of its debugged processes have
21487 terminated in @kbd{target remote} mode. On the other hand, for @kbd{target
21488 extended-remote}, @code{gdbserver} stays running even with no processes left.
21489 @value{GDBN} normally terminates the spawned debugged process on its exit,
21490 which normally also terminates @code{gdbserver} in the @kbd{target remote}
21491 mode. Therefore, when the connection drops unexpectedly, and @value{GDBN}
21492 cannot ask @code{gdbserver} to kill its debugged processes, @code{gdbserver}
21493 stays running even in the @kbd{target remote} mode.
21495 When @code{gdbserver} stays running, @value{GDBN} can connect to it again later.
21496 Such reconnecting is useful for features like @ref{disconnected tracing}. For
21497 completeness, at most one @value{GDBN} can be connected at a time.
21499 @cindex @option{--once}, @code{gdbserver} option
21500 By default, @code{gdbserver} keeps the listening TCP port open, so that
21501 subsequent connections are possible. However, if you start @code{gdbserver}
21502 with the @option{--once} option, it will stop listening for any further
21503 connection attempts after connecting to the first @value{GDBN} session. This
21504 means no further connections to @code{gdbserver} will be possible after the
21505 first one. It also means @code{gdbserver} will terminate after the first
21506 connection with remote @value{GDBN} has closed, even for unexpectedly closed
21507 connections and even in the @kbd{target extended-remote} mode. The
21508 @option{--once} option allows reusing the same port number for connecting to
21509 multiple instances of @code{gdbserver} running on the same host, since each
21510 instance closes its port after the first connection.
21512 @anchor{Other Command-Line Arguments for gdbserver}
21513 @subsubsection Other Command-Line Arguments for @code{gdbserver}
21515 You can use the @option{--multi} option to start @code{gdbserver} without
21516 specifying a program to debug or a process to attach to. Then you can
21517 attach in @code{target extended-remote} mode and run or attach to a
21518 program. For more information,
21519 @pxref{--multi Option in Types of Remote Connnections}.
21521 @cindex @option{--debug}, @code{gdbserver} option
21522 The @option{--debug} option tells @code{gdbserver} to display extra
21523 status information about the debugging process.
21524 @cindex @option{--remote-debug}, @code{gdbserver} option
21525 The @option{--remote-debug} option tells @code{gdbserver} to display
21526 remote protocol debug output.
21527 @cindex @option{--debug-file}, @code{gdbserver} option
21528 @cindex @code{gdbserver}, send all debug output to a single file
21529 The @option{--debug-file=@var{filename}} option tells @code{gdbserver} to
21530 write any debug output to the given @var{filename}. These options are intended
21531 for @code{gdbserver} development and for bug reports to the developers.
21533 @cindex @option{--debug-format}, @code{gdbserver} option
21534 The @option{--debug-format=option1[,option2,...]} option tells
21535 @code{gdbserver} to include additional information in each output.
21536 Possible options are:
21540 Turn off all extra information in debugging output.
21542 Turn on all extra information in debugging output.
21544 Include a timestamp in each line of debugging output.
21547 Options are processed in order. Thus, for example, if @option{none}
21548 appears last then no additional information is added to debugging output.
21550 @cindex @option{--wrapper}, @code{gdbserver} option
21551 The @option{--wrapper} option specifies a wrapper to launch programs
21552 for debugging. The option should be followed by the name of the
21553 wrapper, then any command-line arguments to pass to the wrapper, then
21554 @kbd{--} indicating the end of the wrapper arguments.
21556 @code{gdbserver} runs the specified wrapper program with a combined
21557 command line including the wrapper arguments, then the name of the
21558 program to debug, then any arguments to the program. The wrapper
21559 runs until it executes your program, and then @value{GDBN} gains control.
21561 You can use any program that eventually calls @code{execve} with
21562 its arguments as a wrapper. Several standard Unix utilities do
21563 this, e.g.@: @code{env} and @code{nohup}. Any Unix shell script ending
21564 with @code{exec "$@@"} will also work.
21566 For example, you can use @code{env} to pass an environment variable to
21567 the debugged program, without setting the variable in @code{gdbserver}'s
21571 $ gdbserver --wrapper env LD_PRELOAD=libtest.so -- :2222 ./testprog
21574 @cindex @option{--selftest}
21575 The @option{--selftest} option runs the self tests in @code{gdbserver}:
21578 $ gdbserver --selftest
21579 Ran 2 unit tests, 0 failed
21582 These tests are disabled in release.
21583 @subsection Connecting to @code{gdbserver}
21585 The basic procedure for connecting to the remote target is:
21589 Run @value{GDBN} on the host system.
21592 Make sure you have the necessary symbol files
21593 (@pxref{Host and target files}).
21594 Load symbols for your application using the @code{file} command before you
21595 connect. Use @code{set sysroot} to locate target libraries (unless your
21596 @value{GDBN} was compiled with the correct sysroot using
21597 @code{--with-sysroot}).
21600 Connect to your target (@pxref{Connecting,,Connecting to a Remote Target}).
21601 For TCP connections, you must start up @code{gdbserver} prior to using
21602 the @code{target} command. Otherwise you may get an error whose
21603 text depends on the host system, but which usually looks something like
21604 @samp{Connection refused}. Don't use the @code{load}
21605 command in @value{GDBN} when using @code{target remote} mode, since the
21606 program is already on the target.
21610 @anchor{Monitor Commands for gdbserver}
21611 @subsection Monitor Commands for @code{gdbserver}
21612 @cindex monitor commands, for @code{gdbserver}
21614 During a @value{GDBN} session using @code{gdbserver}, you can use the
21615 @code{monitor} command to send special requests to @code{gdbserver}.
21616 Here are the available commands.
21620 List the available monitor commands.
21622 @item monitor set debug 0
21623 @itemx monitor set debug 1
21624 Disable or enable general debugging messages.
21626 @item monitor set remote-debug 0
21627 @itemx monitor set remote-debug 1
21628 Disable or enable specific debugging messages associated with the remote
21629 protocol (@pxref{Remote Protocol}).
21631 @item monitor set debug-file filename
21632 @itemx monitor set debug-file
21633 Send any debug output to the given file, or to stderr.
21635 @item monitor set debug-format option1@r{[},option2,...@r{]}
21636 Specify additional text to add to debugging messages.
21637 Possible options are:
21641 Turn off all extra information in debugging output.
21643 Turn on all extra information in debugging output.
21645 Include a timestamp in each line of debugging output.
21648 Options are processed in order. Thus, for example, if @option{none}
21649 appears last then no additional information is added to debugging output.
21651 @item monitor set libthread-db-search-path [PATH]
21652 @cindex gdbserver, search path for @code{libthread_db}
21653 When this command is issued, @var{path} is a colon-separated list of
21654 directories to search for @code{libthread_db} (@pxref{Threads,,set
21655 libthread-db-search-path}). If you omit @var{path},
21656 @samp{libthread-db-search-path} will be reset to its default value.
21658 The special entry @samp{$pdir} for @samp{libthread-db-search-path} is
21659 not supported in @code{gdbserver}.
21662 Tell gdbserver to exit immediately. This command should be followed by
21663 @code{disconnect} to close the debugging session. @code{gdbserver} will
21664 detach from any attached processes and kill any processes it created.
21665 Use @code{monitor exit} to terminate @code{gdbserver} at the end
21666 of a multi-process mode debug session.
21670 @subsection Tracepoints support in @code{gdbserver}
21671 @cindex tracepoints support in @code{gdbserver}
21673 On some targets, @code{gdbserver} supports tracepoints, fast
21674 tracepoints and static tracepoints.
21676 For fast or static tracepoints to work, a special library called the
21677 @dfn{in-process agent} (IPA), must be loaded in the inferior process.
21678 This library is built and distributed as an integral part of
21679 @code{gdbserver}. In addition, support for static tracepoints
21680 requires building the in-process agent library with static tracepoints
21681 support. At present, the UST (LTTng Userspace Tracer,
21682 @url{http://lttng.org/ust}) tracing engine is supported. This support
21683 is automatically available if UST development headers are found in the
21684 standard include path when @code{gdbserver} is built, or if
21685 @code{gdbserver} was explicitly configured using @option{--with-ust}
21686 to point at such headers. You can explicitly disable the support
21687 using @option{--with-ust=no}.
21689 There are several ways to load the in-process agent in your program:
21692 @item Specifying it as dependency at link time
21694 You can link your program dynamically with the in-process agent
21695 library. On most systems, this is accomplished by adding
21696 @code{-linproctrace} to the link command.
21698 @item Using the system's preloading mechanisms
21700 You can force loading the in-process agent at startup time by using
21701 your system's support for preloading shared libraries. Many Unixes
21702 support the concept of preloading user defined libraries. In most
21703 cases, you do that by specifying @code{LD_PRELOAD=libinproctrace.so}
21704 in the environment. See also the description of @code{gdbserver}'s
21705 @option{--wrapper} command line option.
21707 @item Using @value{GDBN} to force loading the agent at run time
21709 On some systems, you can force the inferior to load a shared library,
21710 by calling a dynamic loader function in the inferior that takes care
21711 of dynamically looking up and loading a shared library. On most Unix
21712 systems, the function is @code{dlopen}. You'll use the @code{call}
21713 command for that. For example:
21716 (@value{GDBP}) call dlopen ("libinproctrace.so", ...)
21719 Note that on most Unix systems, for the @code{dlopen} function to be
21720 available, the program needs to be linked with @code{-ldl}.
21723 On systems that have a userspace dynamic loader, like most Unix
21724 systems, when you connect to @code{gdbserver} using @code{target
21725 remote}, you'll find that the program is stopped at the dynamic
21726 loader's entry point, and no shared library has been loaded in the
21727 program's address space yet, including the in-process agent. In that
21728 case, before being able to use any of the fast or static tracepoints
21729 features, you need to let the loader run and load the shared
21730 libraries. The simplest way to do that is to run the program to the
21731 main procedure. E.g., if debugging a C or C@t{++} program, start
21732 @code{gdbserver} like so:
21735 $ gdbserver :9999 myprogram
21738 Start GDB and connect to @code{gdbserver} like so, and run to main:
21742 (@value{GDBP}) target remote myhost:9999
21743 0x00007f215893ba60 in ?? () from /lib64/ld-linux-x86-64.so.2
21744 (@value{GDBP}) b main
21745 (@value{GDBP}) continue
21748 The in-process tracing agent library should now be loaded into the
21749 process; you can confirm it with the @code{info sharedlibrary}
21750 command, which will list @file{libinproctrace.so} as loaded in the
21751 process. You are now ready to install fast tracepoints, list static
21752 tracepoint markers, probe static tracepoints markers, and start
21755 @node Remote Configuration
21756 @section Remote Configuration
21759 @kindex show remote
21760 This section documents the configuration options available when
21761 debugging remote programs. For the options related to the File I/O
21762 extensions of the remote protocol, see @ref{system,
21763 system-call-allowed}.
21766 @item set remoteaddresssize @var{bits}
21767 @cindex address size for remote targets
21768 @cindex bits in remote address
21769 Set the maximum size of address in a memory packet to the specified
21770 number of bits. @value{GDBN} will mask off the address bits above
21771 that number, when it passes addresses to the remote target. The
21772 default value is the number of bits in the target's address.
21774 @item show remoteaddresssize
21775 Show the current value of remote address size in bits.
21777 @item set serial baud @var{n}
21778 @cindex baud rate for remote targets
21779 Set the baud rate for the remote serial I/O to @var{n} baud. The
21780 value is used to set the speed of the serial port used for debugging
21783 @item show serial baud
21784 Show the current speed of the remote connection.
21786 @item set serial parity @var{parity}
21787 Set the parity for the remote serial I/O. Supported values of @var{parity} are:
21788 @code{even}, @code{none}, and @code{odd}. The default is @code{none}.
21790 @item show serial parity
21791 Show the current parity of the serial port.
21793 @item set remotebreak
21794 @cindex interrupt remote programs
21795 @cindex BREAK signal instead of Ctrl-C
21796 @anchor{set remotebreak}
21797 If set to on, @value{GDBN} sends a @code{BREAK} signal to the remote
21798 when you type @kbd{Ctrl-c} to interrupt the program running
21799 on the remote. If set to off, @value{GDBN} sends the @samp{Ctrl-C}
21800 character instead. The default is off, since most remote systems
21801 expect to see @samp{Ctrl-C} as the interrupt signal.
21803 @item show remotebreak
21804 Show whether @value{GDBN} sends @code{BREAK} or @samp{Ctrl-C} to
21805 interrupt the remote program.
21807 @item set remoteflow on
21808 @itemx set remoteflow off
21809 @kindex set remoteflow
21810 Enable or disable hardware flow control (@code{RTS}/@code{CTS})
21811 on the serial port used to communicate to the remote target.
21813 @item show remoteflow
21814 @kindex show remoteflow
21815 Show the current setting of hardware flow control.
21817 @item set remotelogbase @var{base}
21818 Set the base (a.k.a.@: radix) of logging serial protocol
21819 communications to @var{base}. Supported values of @var{base} are:
21820 @code{ascii}, @code{octal}, and @code{hex}. The default is
21823 @item show remotelogbase
21824 Show the current setting of the radix for logging remote serial
21827 @item set remotelogfile @var{file}
21828 @cindex record serial communications on file
21829 Record remote serial communications on the named @var{file}. The
21830 default is not to record at all.
21832 @item show remotelogfile
21833 Show the current setting of the file name on which to record the
21834 serial communications.
21836 @item set remotetimeout @var{num}
21837 @cindex timeout for serial communications
21838 @cindex remote timeout
21839 Set the timeout limit to wait for the remote target to respond to
21840 @var{num} seconds. The default is 2 seconds.
21842 @item show remotetimeout
21843 Show the current number of seconds to wait for the remote target
21846 @cindex limit hardware breakpoints and watchpoints
21847 @cindex remote target, limit break- and watchpoints
21848 @anchor{set remote hardware-watchpoint-limit}
21849 @anchor{set remote hardware-breakpoint-limit}
21850 @item set remote hardware-watchpoint-limit @var{limit}
21851 @itemx set remote hardware-breakpoint-limit @var{limit}
21852 Restrict @value{GDBN} to using @var{limit} remote hardware watchpoints
21853 or breakpoints. The @var{limit} can be set to 0 to disable hardware
21854 watchpoints or breakpoints, and @code{unlimited} for unlimited
21855 watchpoints or breakpoints.
21857 @item show remote hardware-watchpoint-limit
21858 @itemx show remote hardware-breakpoint-limit
21859 Show the current limit for the number of hardware watchpoints or
21860 breakpoints that @value{GDBN} can use.
21862 @cindex limit hardware watchpoints length
21863 @cindex remote target, limit watchpoints length
21864 @anchor{set remote hardware-watchpoint-length-limit}
21865 @item set remote hardware-watchpoint-length-limit @var{limit}
21866 Restrict @value{GDBN} to using @var{limit} bytes for the maximum
21867 length of a remote hardware watchpoint. A @var{limit} of 0 disables
21868 hardware watchpoints and @code{unlimited} allows watchpoints of any
21871 @item show remote hardware-watchpoint-length-limit
21872 Show the current limit (in bytes) of the maximum length of
21873 a remote hardware watchpoint.
21875 @item set remote exec-file @var{filename}
21876 @itemx show remote exec-file
21877 @anchor{set remote exec-file}
21878 @cindex executable file, for remote target
21879 Select the file used for @code{run} with @code{target
21880 extended-remote}. This should be set to a filename valid on the
21881 target system. If it is not set, the target will use a default
21882 filename (e.g.@: the last program run).
21884 @item set remote interrupt-sequence
21885 @cindex interrupt remote programs
21886 @cindex select Ctrl-C, BREAK or BREAK-g
21887 Allow the user to select one of @samp{Ctrl-C}, a @code{BREAK} or
21888 @samp{BREAK-g} as the
21889 sequence to the remote target in order to interrupt the execution.
21890 @samp{Ctrl-C} is a default. Some system prefers @code{BREAK} which
21891 is high level of serial line for some certain time.
21892 Linux kernel prefers @samp{BREAK-g}, a.k.a Magic SysRq g.
21893 It is @code{BREAK} signal followed by character @code{g}.
21895 @item show interrupt-sequence
21896 Show which of @samp{Ctrl-C}, @code{BREAK} or @code{BREAK-g}
21897 is sent by @value{GDBN} to interrupt the remote program.
21898 @code{BREAK-g} is BREAK signal followed by @code{g} and
21899 also known as Magic SysRq g.
21901 @item set remote interrupt-on-connect
21902 @cindex send interrupt-sequence on start
21903 Specify whether interrupt-sequence is sent to remote target when
21904 @value{GDBN} connects to it. This is mostly needed when you debug
21905 Linux kernel. Linux kernel expects @code{BREAK} followed by @code{g}
21906 which is known as Magic SysRq g in order to connect @value{GDBN}.
21908 @item show interrupt-on-connect
21909 Show whether interrupt-sequence is sent
21910 to remote target when @value{GDBN} connects to it.
21914 @item set tcp auto-retry on
21915 @cindex auto-retry, for remote TCP target
21916 Enable auto-retry for remote TCP connections. This is useful if the remote
21917 debugging agent is launched in parallel with @value{GDBN}; there is a race
21918 condition because the agent may not become ready to accept the connection
21919 before @value{GDBN} attempts to connect. When auto-retry is
21920 enabled, if the initial attempt to connect fails, @value{GDBN} reattempts
21921 to establish the connection using the timeout specified by
21922 @code{set tcp connect-timeout}.
21924 @item set tcp auto-retry off
21925 Do not auto-retry failed TCP connections.
21927 @item show tcp auto-retry
21928 Show the current auto-retry setting.
21930 @item set tcp connect-timeout @var{seconds}
21931 @itemx set tcp connect-timeout unlimited
21932 @cindex connection timeout, for remote TCP target
21933 @cindex timeout, for remote target connection
21934 Set the timeout for establishing a TCP connection to the remote target to
21935 @var{seconds}. The timeout affects both polling to retry failed connections
21936 (enabled by @code{set tcp auto-retry on}) and waiting for connections
21937 that are merely slow to complete, and represents an approximate cumulative
21938 value. If @var{seconds} is @code{unlimited}, there is no timeout and
21939 @value{GDBN} will keep attempting to establish a connection forever,
21940 unless interrupted with @kbd{Ctrl-c}. The default is 15 seconds.
21942 @item show tcp connect-timeout
21943 Show the current connection timeout setting.
21946 @cindex remote packets, enabling and disabling
21947 The @value{GDBN} remote protocol autodetects the packets supported by
21948 your debugging stub. If you need to override the autodetection, you
21949 can use these commands to enable or disable individual packets. Each
21950 packet can be set to @samp{on} (the remote target supports this
21951 packet), @samp{off} (the remote target does not support this packet),
21952 or @samp{auto} (detect remote target support for this packet). They
21953 all default to @samp{auto}. For more information about each packet,
21954 see @ref{Remote Protocol}.
21956 During normal use, you should not have to use any of these commands.
21957 If you do, that may be a bug in your remote debugging stub, or a bug
21958 in @value{GDBN}. You may want to report the problem to the
21959 @value{GDBN} developers.
21961 For each packet @var{name}, the command to enable or disable the
21962 packet is @code{set remote @var{name}-packet}. The available settings
21965 @multitable @columnfractions 0.28 0.32 0.25
21968 @tab Related Features
21970 @item @code{fetch-register}
21972 @tab @code{info registers}
21974 @item @code{set-register}
21978 @item @code{binary-download}
21980 @tab @code{load}, @code{set}
21982 @item @code{read-aux-vector}
21983 @tab @code{qXfer:auxv:read}
21984 @tab @code{info auxv}
21986 @item @code{symbol-lookup}
21987 @tab @code{qSymbol}
21988 @tab Detecting multiple threads
21990 @item @code{attach}
21991 @tab @code{vAttach}
21994 @item @code{verbose-resume}
21996 @tab Stepping or resuming multiple threads
22002 @item @code{software-breakpoint}
22006 @item @code{hardware-breakpoint}
22010 @item @code{write-watchpoint}
22014 @item @code{read-watchpoint}
22018 @item @code{access-watchpoint}
22022 @item @code{pid-to-exec-file}
22023 @tab @code{qXfer:exec-file:read}
22024 @tab @code{attach}, @code{run}
22026 @item @code{target-features}
22027 @tab @code{qXfer:features:read}
22028 @tab @code{set architecture}
22030 @item @code{library-info}
22031 @tab @code{qXfer:libraries:read}
22032 @tab @code{info sharedlibrary}
22034 @item @code{memory-map}
22035 @tab @code{qXfer:memory-map:read}
22036 @tab @code{info mem}
22038 @item @code{read-sdata-object}
22039 @tab @code{qXfer:sdata:read}
22040 @tab @code{print $_sdata}
22042 @item @code{read-spu-object}
22043 @tab @code{qXfer:spu:read}
22044 @tab @code{info spu}
22046 @item @code{write-spu-object}
22047 @tab @code{qXfer:spu:write}
22048 @tab @code{info spu}
22050 @item @code{read-siginfo-object}
22051 @tab @code{qXfer:siginfo:read}
22052 @tab @code{print $_siginfo}
22054 @item @code{write-siginfo-object}
22055 @tab @code{qXfer:siginfo:write}
22056 @tab @code{set $_siginfo}
22058 @item @code{threads}
22059 @tab @code{qXfer:threads:read}
22060 @tab @code{info threads}
22062 @item @code{get-thread-local-@*storage-address}
22063 @tab @code{qGetTLSAddr}
22064 @tab Displaying @code{__thread} variables
22066 @item @code{get-thread-information-block-address}
22067 @tab @code{qGetTIBAddr}
22068 @tab Display MS-Windows Thread Information Block.
22070 @item @code{search-memory}
22071 @tab @code{qSearch:memory}
22074 @item @code{supported-packets}
22075 @tab @code{qSupported}
22076 @tab Remote communications parameters
22078 @item @code{catch-syscalls}
22079 @tab @code{QCatchSyscalls}
22080 @tab @code{catch syscall}
22082 @item @code{pass-signals}
22083 @tab @code{QPassSignals}
22084 @tab @code{handle @var{signal}}
22086 @item @code{program-signals}
22087 @tab @code{QProgramSignals}
22088 @tab @code{handle @var{signal}}
22090 @item @code{hostio-close-packet}
22091 @tab @code{vFile:close}
22092 @tab @code{remote get}, @code{remote put}
22094 @item @code{hostio-open-packet}
22095 @tab @code{vFile:open}
22096 @tab @code{remote get}, @code{remote put}
22098 @item @code{hostio-pread-packet}
22099 @tab @code{vFile:pread}
22100 @tab @code{remote get}, @code{remote put}
22102 @item @code{hostio-pwrite-packet}
22103 @tab @code{vFile:pwrite}
22104 @tab @code{remote get}, @code{remote put}
22106 @item @code{hostio-unlink-packet}
22107 @tab @code{vFile:unlink}
22108 @tab @code{remote delete}
22110 @item @code{hostio-readlink-packet}
22111 @tab @code{vFile:readlink}
22114 @item @code{hostio-fstat-packet}
22115 @tab @code{vFile:fstat}
22118 @item @code{hostio-setfs-packet}
22119 @tab @code{vFile:setfs}
22122 @item @code{noack-packet}
22123 @tab @code{QStartNoAckMode}
22124 @tab Packet acknowledgment
22126 @item @code{osdata}
22127 @tab @code{qXfer:osdata:read}
22128 @tab @code{info os}
22130 @item @code{query-attached}
22131 @tab @code{qAttached}
22132 @tab Querying remote process attach state.
22134 @item @code{trace-buffer-size}
22135 @tab @code{QTBuffer:size}
22136 @tab @code{set trace-buffer-size}
22138 @item @code{trace-status}
22139 @tab @code{qTStatus}
22140 @tab @code{tstatus}
22142 @item @code{traceframe-info}
22143 @tab @code{qXfer:traceframe-info:read}
22144 @tab Traceframe info
22146 @item @code{install-in-trace}
22147 @tab @code{InstallInTrace}
22148 @tab Install tracepoint in tracing
22150 @item @code{disable-randomization}
22151 @tab @code{QDisableRandomization}
22152 @tab @code{set disable-randomization}
22154 @item @code{startup-with-shell}
22155 @tab @code{QStartupWithShell}
22156 @tab @code{set startup-with-shell}
22158 @item @code{environment-hex-encoded}
22159 @tab @code{QEnvironmentHexEncoded}
22160 @tab @code{set environment}
22162 @item @code{environment-unset}
22163 @tab @code{QEnvironmentUnset}
22164 @tab @code{unset environment}
22166 @item @code{environment-reset}
22167 @tab @code{QEnvironmentReset}
22168 @tab @code{Reset the inferior environment (i.e., unset user-set variables)}
22170 @item @code{set-working-dir}
22171 @tab @code{QSetWorkingDir}
22172 @tab @code{set cwd}
22174 @item @code{conditional-breakpoints-packet}
22175 @tab @code{Z0 and Z1}
22176 @tab @code{Support for target-side breakpoint condition evaluation}
22178 @item @code{multiprocess-extensions}
22179 @tab @code{multiprocess extensions}
22180 @tab Debug multiple processes and remote process PID awareness
22182 @item @code{swbreak-feature}
22183 @tab @code{swbreak stop reason}
22186 @item @code{hwbreak-feature}
22187 @tab @code{hwbreak stop reason}
22190 @item @code{fork-event-feature}
22191 @tab @code{fork stop reason}
22194 @item @code{vfork-event-feature}
22195 @tab @code{vfork stop reason}
22198 @item @code{exec-event-feature}
22199 @tab @code{exec stop reason}
22202 @item @code{thread-events}
22203 @tab @code{QThreadEvents}
22204 @tab Tracking thread lifetime.
22206 @item @code{no-resumed-stop-reply}
22207 @tab @code{no resumed thread left stop reply}
22208 @tab Tracking thread lifetime.
22213 @section Implementing a Remote Stub
22215 @cindex debugging stub, example
22216 @cindex remote stub, example
22217 @cindex stub example, remote debugging
22218 The stub files provided with @value{GDBN} implement the target side of the
22219 communication protocol, and the @value{GDBN} side is implemented in the
22220 @value{GDBN} source file @file{remote.c}. Normally, you can simply allow
22221 these subroutines to communicate, and ignore the details. (If you're
22222 implementing your own stub file, you can still ignore the details: start
22223 with one of the existing stub files. @file{sparc-stub.c} is the best
22224 organized, and therefore the easiest to read.)
22226 @cindex remote serial debugging, overview
22227 To debug a program running on another machine (the debugging
22228 @dfn{target} machine), you must first arrange for all the usual
22229 prerequisites for the program to run by itself. For example, for a C
22234 A startup routine to set up the C runtime environment; these usually
22235 have a name like @file{crt0}. The startup routine may be supplied by
22236 your hardware supplier, or you may have to write your own.
22239 A C subroutine library to support your program's
22240 subroutine calls, notably managing input and output.
22243 A way of getting your program to the other machine---for example, a
22244 download program. These are often supplied by the hardware
22245 manufacturer, but you may have to write your own from hardware
22249 The next step is to arrange for your program to use a serial port to
22250 communicate with the machine where @value{GDBN} is running (the @dfn{host}
22251 machine). In general terms, the scheme looks like this:
22255 @value{GDBN} already understands how to use this protocol; when everything
22256 else is set up, you can simply use the @samp{target remote} command
22257 (@pxref{Targets,,Specifying a Debugging Target}).
22259 @item On the target,
22260 you must link with your program a few special-purpose subroutines that
22261 implement the @value{GDBN} remote serial protocol. The file containing these
22262 subroutines is called a @dfn{debugging stub}.
22264 On certain remote targets, you can use an auxiliary program
22265 @code{gdbserver} instead of linking a stub into your program.
22266 @xref{Server,,Using the @code{gdbserver} Program}, for details.
22269 The debugging stub is specific to the architecture of the remote
22270 machine; for example, use @file{sparc-stub.c} to debug programs on
22273 @cindex remote serial stub list
22274 These working remote stubs are distributed with @value{GDBN}:
22279 @cindex @file{i386-stub.c}
22282 For Intel 386 and compatible architectures.
22285 @cindex @file{m68k-stub.c}
22286 @cindex Motorola 680x0
22288 For Motorola 680x0 architectures.
22291 @cindex @file{sh-stub.c}
22294 For Renesas SH architectures.
22297 @cindex @file{sparc-stub.c}
22299 For @sc{sparc} architectures.
22301 @item sparcl-stub.c
22302 @cindex @file{sparcl-stub.c}
22305 For Fujitsu @sc{sparclite} architectures.
22309 The @file{README} file in the @value{GDBN} distribution may list other
22310 recently added stubs.
22313 * Stub Contents:: What the stub can do for you
22314 * Bootstrapping:: What you must do for the stub
22315 * Debug Session:: Putting it all together
22318 @node Stub Contents
22319 @subsection What the Stub Can Do for You
22321 @cindex remote serial stub
22322 The debugging stub for your architecture supplies these three
22326 @item set_debug_traps
22327 @findex set_debug_traps
22328 @cindex remote serial stub, initialization
22329 This routine arranges for @code{handle_exception} to run when your
22330 program stops. You must call this subroutine explicitly in your
22331 program's startup code.
22333 @item handle_exception
22334 @findex handle_exception
22335 @cindex remote serial stub, main routine
22336 This is the central workhorse, but your program never calls it
22337 explicitly---the setup code arranges for @code{handle_exception} to
22338 run when a trap is triggered.
22340 @code{handle_exception} takes control when your program stops during
22341 execution (for example, on a breakpoint), and mediates communications
22342 with @value{GDBN} on the host machine. This is where the communications
22343 protocol is implemented; @code{handle_exception} acts as the @value{GDBN}
22344 representative on the target machine. It begins by sending summary
22345 information on the state of your program, then continues to execute,
22346 retrieving and transmitting any information @value{GDBN} needs, until you
22347 execute a @value{GDBN} command that makes your program resume; at that point,
22348 @code{handle_exception} returns control to your own code on the target
22352 @cindex @code{breakpoint} subroutine, remote
22353 Use this auxiliary subroutine to make your program contain a
22354 breakpoint. Depending on the particular situation, this may be the only
22355 way for @value{GDBN} to get control. For instance, if your target
22356 machine has some sort of interrupt button, you won't need to call this;
22357 pressing the interrupt button transfers control to
22358 @code{handle_exception}---in effect, to @value{GDBN}. On some machines,
22359 simply receiving characters on the serial port may also trigger a trap;
22360 again, in that situation, you don't need to call @code{breakpoint} from
22361 your own program---simply running @samp{target remote} from the host
22362 @value{GDBN} session gets control.
22364 Call @code{breakpoint} if none of these is true, or if you simply want
22365 to make certain your program stops at a predetermined point for the
22366 start of your debugging session.
22369 @node Bootstrapping
22370 @subsection What You Must Do for the Stub
22372 @cindex remote stub, support routines
22373 The debugging stubs that come with @value{GDBN} are set up for a particular
22374 chip architecture, but they have no information about the rest of your
22375 debugging target machine.
22377 First of all you need to tell the stub how to communicate with the
22381 @item int getDebugChar()
22382 @findex getDebugChar
22383 Write this subroutine to read a single character from the serial port.
22384 It may be identical to @code{getchar} for your target system; a
22385 different name is used to allow you to distinguish the two if you wish.
22387 @item void putDebugChar(int)
22388 @findex putDebugChar
22389 Write this subroutine to write a single character to the serial port.
22390 It may be identical to @code{putchar} for your target system; a
22391 different name is used to allow you to distinguish the two if you wish.
22394 @cindex control C, and remote debugging
22395 @cindex interrupting remote targets
22396 If you want @value{GDBN} to be able to stop your program while it is
22397 running, you need to use an interrupt-driven serial driver, and arrange
22398 for it to stop when it receives a @code{^C} (@samp{\003}, the control-C
22399 character). That is the character which @value{GDBN} uses to tell the
22400 remote system to stop.
22402 Getting the debugging target to return the proper status to @value{GDBN}
22403 probably requires changes to the standard stub; one quick and dirty way
22404 is to just execute a breakpoint instruction (the ``dirty'' part is that
22405 @value{GDBN} reports a @code{SIGTRAP} instead of a @code{SIGINT}).
22407 Other routines you need to supply are:
22410 @item void exceptionHandler (int @var{exception_number}, void *@var{exception_address})
22411 @findex exceptionHandler
22412 Write this function to install @var{exception_address} in the exception
22413 handling tables. You need to do this because the stub does not have any
22414 way of knowing what the exception handling tables on your target system
22415 are like (for example, the processor's table might be in @sc{rom},
22416 containing entries which point to a table in @sc{ram}).
22417 The @var{exception_number} specifies the exception which should be changed;
22418 its meaning is architecture-dependent (for example, different numbers
22419 might represent divide by zero, misaligned access, etc). When this
22420 exception occurs, control should be transferred directly to
22421 @var{exception_address}, and the processor state (stack, registers,
22422 and so on) should be just as it is when a processor exception occurs. So if
22423 you want to use a jump instruction to reach @var{exception_address}, it
22424 should be a simple jump, not a jump to subroutine.
22426 For the 386, @var{exception_address} should be installed as an interrupt
22427 gate so that interrupts are masked while the handler runs. The gate
22428 should be at privilege level 0 (the most privileged level). The
22429 @sc{sparc} and 68k stubs are able to mask interrupts themselves without
22430 help from @code{exceptionHandler}.
22432 @item void flush_i_cache()
22433 @findex flush_i_cache
22434 On @sc{sparc} and @sc{sparclite} only, write this subroutine to flush the
22435 instruction cache, if any, on your target machine. If there is no
22436 instruction cache, this subroutine may be a no-op.
22438 On target machines that have instruction caches, @value{GDBN} requires this
22439 function to make certain that the state of your program is stable.
22443 You must also make sure this library routine is available:
22446 @item void *memset(void *, int, int)
22448 This is the standard library function @code{memset} that sets an area of
22449 memory to a known value. If you have one of the free versions of
22450 @code{libc.a}, @code{memset} can be found there; otherwise, you must
22451 either obtain it from your hardware manufacturer, or write your own.
22454 If you do not use the GNU C compiler, you may need other standard
22455 library subroutines as well; this varies from one stub to another,
22456 but in general the stubs are likely to use any of the common library
22457 subroutines which @code{@value{NGCC}} generates as inline code.
22460 @node Debug Session
22461 @subsection Putting it All Together
22463 @cindex remote serial debugging summary
22464 In summary, when your program is ready to debug, you must follow these
22469 Make sure you have defined the supporting low-level routines
22470 (@pxref{Bootstrapping,,What You Must Do for the Stub}):
22472 @code{getDebugChar}, @code{putDebugChar},
22473 @code{flush_i_cache}, @code{memset}, @code{exceptionHandler}.
22477 Insert these lines in your program's startup code, before the main
22478 procedure is called:
22485 On some machines, when a breakpoint trap is raised, the hardware
22486 automatically makes the PC point to the instruction after the
22487 breakpoint. If your machine doesn't do that, you may need to adjust
22488 @code{handle_exception} to arrange for it to return to the instruction
22489 after the breakpoint on this first invocation, so that your program
22490 doesn't keep hitting the initial breakpoint instead of making
22494 For the 680x0 stub only, you need to provide a variable called
22495 @code{exceptionHook}. Normally you just use:
22498 void (*exceptionHook)() = 0;
22502 but if before calling @code{set_debug_traps}, you set it to point to a
22503 function in your program, that function is called when
22504 @code{@value{GDBN}} continues after stopping on a trap (for example, bus
22505 error). The function indicated by @code{exceptionHook} is called with
22506 one parameter: an @code{int} which is the exception number.
22509 Compile and link together: your program, the @value{GDBN} debugging stub for
22510 your target architecture, and the supporting subroutines.
22513 Make sure you have a serial connection between your target machine and
22514 the @value{GDBN} host, and identify the serial port on the host.
22517 @c The "remote" target now provides a `load' command, so we should
22518 @c document that. FIXME.
22519 Download your program to your target machine (or get it there by
22520 whatever means the manufacturer provides), and start it.
22523 Start @value{GDBN} on the host, and connect to the target
22524 (@pxref{Connecting,,Connecting to a Remote Target}).
22528 @node Configurations
22529 @chapter Configuration-Specific Information
22531 While nearly all @value{GDBN} commands are available for all native and
22532 cross versions of the debugger, there are some exceptions. This chapter
22533 describes things that are only available in certain configurations.
22535 There are three major categories of configurations: native
22536 configurations, where the host and target are the same, embedded
22537 operating system configurations, which are usually the same for several
22538 different processor architectures, and bare embedded processors, which
22539 are quite different from each other.
22544 * Embedded Processors::
22551 This section describes details specific to particular native
22555 * BSD libkvm Interface:: Debugging BSD kernel memory images
22556 * Process Information:: Process information
22557 * DJGPP Native:: Features specific to the DJGPP port
22558 * Cygwin Native:: Features specific to the Cygwin port
22559 * Hurd Native:: Features specific to @sc{gnu} Hurd
22560 * Darwin:: Features specific to Darwin
22561 * FreeBSD:: Features specific to FreeBSD
22564 @node BSD libkvm Interface
22565 @subsection BSD libkvm Interface
22568 @cindex kernel memory image
22569 @cindex kernel crash dump
22571 BSD-derived systems (FreeBSD/NetBSD/OpenBSD) have a kernel memory
22572 interface that provides a uniform interface for accessing kernel virtual
22573 memory images, including live systems and crash dumps. @value{GDBN}
22574 uses this interface to allow you to debug live kernels and kernel crash
22575 dumps on many native BSD configurations. This is implemented as a
22576 special @code{kvm} debugging target. For debugging a live system, load
22577 the currently running kernel into @value{GDBN} and connect to the
22581 (@value{GDBP}) @b{target kvm}
22584 For debugging crash dumps, provide the file name of the crash dump as an
22588 (@value{GDBP}) @b{target kvm /var/crash/bsd.0}
22591 Once connected to the @code{kvm} target, the following commands are
22597 Set current context from the @dfn{Process Control Block} (PCB) address.
22600 Set current context from proc address. This command isn't available on
22601 modern FreeBSD systems.
22604 @node Process Information
22605 @subsection Process Information
22607 @cindex examine process image
22608 @cindex process info via @file{/proc}
22610 Some operating systems provide interfaces to fetch additional
22611 information about running processes beyond memory and per-thread
22612 register state. If @value{GDBN} is configured for an operating system
22613 with a supported interface, the command @code{info proc} is available
22614 to report information about the process running your program, or about
22615 any process running on your system.
22617 One supported interface is a facility called @samp{/proc} that can be
22618 used to examine the image of a running process using file-system
22619 subroutines. This facility is supported on @sc{gnu}/Linux and Solaris
22622 On FreeBSD systems, system control nodes are used to query process
22625 In addition, some systems may provide additional process information
22626 in core files. Note that a core file may include a subset of the
22627 information available from a live process. Process information is
22628 currently avaiable from cores created on @sc{gnu}/Linux and FreeBSD
22635 @itemx info proc @var{process-id}
22636 Summarize available information about a process. If a
22637 process ID is specified by @var{process-id}, display information about
22638 that process; otherwise display information about the program being
22639 debugged. The summary includes the debugged process ID, the command
22640 line used to invoke it, its current working directory, and its
22641 executable file's absolute file name.
22643 On some systems, @var{process-id} can be of the form
22644 @samp{[@var{pid}]/@var{tid}} which specifies a certain thread ID
22645 within a process. If the optional @var{pid} part is missing, it means
22646 a thread from the process being debugged (the leading @samp{/} still
22647 needs to be present, or else @value{GDBN} will interpret the number as
22648 a process ID rather than a thread ID).
22650 @item info proc cmdline
22651 @cindex info proc cmdline
22652 Show the original command line of the process. This command is
22653 supported on @sc{gnu}/Linux and FreeBSD.
22655 @item info proc cwd
22656 @cindex info proc cwd
22657 Show the current working directory of the process. This command is
22658 supported on @sc{gnu}/Linux and FreeBSD.
22660 @item info proc exe
22661 @cindex info proc exe
22662 Show the name of executable of the process. This command is supported
22663 on @sc{gnu}/Linux and FreeBSD.
22665 @item info proc files
22666 @cindex info proc files
22667 Show the file descriptors open by the process. For each open file
22668 descriptor, @value{GDBN} shows its number, type (file, directory,
22669 character device, socket), file pointer offset, and the name of the
22670 resource open on the descriptor. The resource name can be a file name
22671 (for files, directories, and devices) or a protocol followed by socket
22672 address (for network connections). This command is supported on
22675 This example shows the open file descriptors for a process using a
22676 tty for standard input and output as well as two network sockets:
22679 (gdb) info proc files 22136
22683 FD Type Offset Flags Name
22684 text file - r-------- /usr/bin/ssh
22685 ctty chr - rw------- /dev/pts/20
22686 cwd dir - r-------- /usr/home/john
22687 root dir - r-------- /
22688 0 chr 0x32933a4 rw------- /dev/pts/20
22689 1 chr 0x32933a4 rw------- /dev/pts/20
22690 2 chr 0x32933a4 rw------- /dev/pts/20
22691 3 socket 0x0 rw----n-- tcp4 10.0.1.2:53014 -> 10.0.1.10:22
22692 4 socket 0x0 rw------- unix stream:/tmp/ssh-FIt89oAzOn5f/agent.2456
22695 @item info proc mappings
22696 @cindex memory address space mappings
22697 Report the memory address space ranges accessible in a process. On
22698 Solaris and FreeBSD systems, each memory range includes information on
22699 whether the process has read, write, or execute access rights to each
22700 range. On @sc{gnu}/Linux and FreeBSD systems, each memory range
22701 includes the object file which is mapped to that range.
22703 @item info proc stat
22704 @itemx info proc status
22705 @cindex process detailed status information
22706 Show additional process-related information, including the user ID and
22707 group ID; virtual memory usage; the signals that are pending, blocked,
22708 and ignored; its TTY; its consumption of system and user time; its
22709 stack size; its @samp{nice} value; etc. These commands are supported
22710 on @sc{gnu}/Linux and FreeBSD.
22712 For @sc{gnu}/Linux systems, see the @samp{proc} man page for more
22713 information (type @kbd{man 5 proc} from your shell prompt).
22715 For FreeBSD systems, @code{info proc stat} is an alias for @code{info
22718 @item info proc all
22719 Show all the information about the process described under all of the
22720 above @code{info proc} subcommands.
22723 @comment These sub-options of 'info proc' were not included when
22724 @comment procfs.c was re-written. Keep their descriptions around
22725 @comment against the day when someone finds the time to put them back in.
22726 @kindex info proc times
22727 @item info proc times
22728 Starting time, user CPU time, and system CPU time for your program and
22731 @kindex info proc id
22733 Report on the process IDs related to your program: its own process ID,
22734 the ID of its parent, the process group ID, and the session ID.
22737 @item set procfs-trace
22738 @kindex set procfs-trace
22739 @cindex @code{procfs} API calls
22740 This command enables and disables tracing of @code{procfs} API calls.
22742 @item show procfs-trace
22743 @kindex show procfs-trace
22744 Show the current state of @code{procfs} API call tracing.
22746 @item set procfs-file @var{file}
22747 @kindex set procfs-file
22748 Tell @value{GDBN} to write @code{procfs} API trace to the named
22749 @var{file}. @value{GDBN} appends the trace info to the previous
22750 contents of the file. The default is to display the trace on the
22753 @item show procfs-file
22754 @kindex show procfs-file
22755 Show the file to which @code{procfs} API trace is written.
22757 @item proc-trace-entry
22758 @itemx proc-trace-exit
22759 @itemx proc-untrace-entry
22760 @itemx proc-untrace-exit
22761 @kindex proc-trace-entry
22762 @kindex proc-trace-exit
22763 @kindex proc-untrace-entry
22764 @kindex proc-untrace-exit
22765 These commands enable and disable tracing of entries into and exits
22766 from the @code{syscall} interface.
22769 @kindex info pidlist
22770 @cindex process list, QNX Neutrino
22771 For QNX Neutrino only, this command displays the list of all the
22772 processes and all the threads within each process.
22775 @kindex info meminfo
22776 @cindex mapinfo list, QNX Neutrino
22777 For QNX Neutrino only, this command displays the list of all mapinfos.
22781 @subsection Features for Debugging @sc{djgpp} Programs
22782 @cindex @sc{djgpp} debugging
22783 @cindex native @sc{djgpp} debugging
22784 @cindex MS-DOS-specific commands
22787 @sc{djgpp} is a port of the @sc{gnu} development tools to MS-DOS and
22788 MS-Windows. @sc{djgpp} programs are 32-bit protected-mode programs
22789 that use the @dfn{DPMI} (DOS Protected-Mode Interface) API to run on
22790 top of real-mode DOS systems and their emulations.
22792 @value{GDBN} supports native debugging of @sc{djgpp} programs, and
22793 defines a few commands specific to the @sc{djgpp} port. This
22794 subsection describes those commands.
22799 This is a prefix of @sc{djgpp}-specific commands which print
22800 information about the target system and important OS structures.
22803 @cindex MS-DOS system info
22804 @cindex free memory information (MS-DOS)
22805 @item info dos sysinfo
22806 This command displays assorted information about the underlying
22807 platform: the CPU type and features, the OS version and flavor, the
22808 DPMI version, and the available conventional and DPMI memory.
22813 @cindex segment descriptor tables
22814 @cindex descriptor tables display
22816 @itemx info dos ldt
22817 @itemx info dos idt
22818 These 3 commands display entries from, respectively, Global, Local,
22819 and Interrupt Descriptor Tables (GDT, LDT, and IDT). The descriptor
22820 tables are data structures which store a descriptor for each segment
22821 that is currently in use. The segment's selector is an index into a
22822 descriptor table; the table entry for that index holds the
22823 descriptor's base address and limit, and its attributes and access
22826 A typical @sc{djgpp} program uses 3 segments: a code segment, a data
22827 segment (used for both data and the stack), and a DOS segment (which
22828 allows access to DOS/BIOS data structures and absolute addresses in
22829 conventional memory). However, the DPMI host will usually define
22830 additional segments in order to support the DPMI environment.
22832 @cindex garbled pointers
22833 These commands allow to display entries from the descriptor tables.
22834 Without an argument, all entries from the specified table are
22835 displayed. An argument, which should be an integer expression, means
22836 display a single entry whose index is given by the argument. For
22837 example, here's a convenient way to display information about the
22838 debugged program's data segment:
22841 @exdent @code{(@value{GDBP}) info dos ldt $ds}
22842 @exdent @code{0x13f: base=0x11970000 limit=0x0009ffff 32-Bit Data (Read/Write, Exp-up)}
22846 This comes in handy when you want to see whether a pointer is outside
22847 the data segment's limit (i.e.@: @dfn{garbled}).
22849 @cindex page tables display (MS-DOS)
22851 @itemx info dos pte
22852 These two commands display entries from, respectively, the Page
22853 Directory and the Page Tables. Page Directories and Page Tables are
22854 data structures which control how virtual memory addresses are mapped
22855 into physical addresses. A Page Table includes an entry for every
22856 page of memory that is mapped into the program's address space; there
22857 may be several Page Tables, each one holding up to 4096 entries. A
22858 Page Directory has up to 4096 entries, one each for every Page Table
22859 that is currently in use.
22861 Without an argument, @kbd{info dos pde} displays the entire Page
22862 Directory, and @kbd{info dos pte} displays all the entries in all of
22863 the Page Tables. An argument, an integer expression, given to the
22864 @kbd{info dos pde} command means display only that entry from the Page
22865 Directory table. An argument given to the @kbd{info dos pte} command
22866 means display entries from a single Page Table, the one pointed to by
22867 the specified entry in the Page Directory.
22869 @cindex direct memory access (DMA) on MS-DOS
22870 These commands are useful when your program uses @dfn{DMA} (Direct
22871 Memory Access), which needs physical addresses to program the DMA
22874 These commands are supported only with some DPMI servers.
22876 @cindex physical address from linear address
22877 @item info dos address-pte @var{addr}
22878 This command displays the Page Table entry for a specified linear
22879 address. The argument @var{addr} is a linear address which should
22880 already have the appropriate segment's base address added to it,
22881 because this command accepts addresses which may belong to @emph{any}
22882 segment. For example, here's how to display the Page Table entry for
22883 the page where a variable @code{i} is stored:
22886 @exdent @code{(@value{GDBP}) info dos address-pte __djgpp_base_address + (char *)&i}
22887 @exdent @code{Page Table entry for address 0x11a00d30:}
22888 @exdent @code{Base=0x02698000 Dirty Acc. Not-Cached Write-Back Usr Read-Write +0xd30}
22892 This says that @code{i} is stored at offset @code{0xd30} from the page
22893 whose physical base address is @code{0x02698000}, and shows all the
22894 attributes of that page.
22896 Note that you must cast the addresses of variables to a @code{char *},
22897 since otherwise the value of @code{__djgpp_base_address}, the base
22898 address of all variables and functions in a @sc{djgpp} program, will
22899 be added using the rules of C pointer arithmetics: if @code{i} is
22900 declared an @code{int}, @value{GDBN} will add 4 times the value of
22901 @code{__djgpp_base_address} to the address of @code{i}.
22903 Here's another example, it displays the Page Table entry for the
22907 @exdent @code{(@value{GDBP}) info dos address-pte *((unsigned *)&_go32_info_block + 3)}
22908 @exdent @code{Page Table entry for address 0x29110:}
22909 @exdent @code{Base=0x00029000 Dirty Acc. Not-Cached Write-Back Usr Read-Write +0x110}
22913 (The @code{+ 3} offset is because the transfer buffer's address is the
22914 3rd member of the @code{_go32_info_block} structure.) The output
22915 clearly shows that this DPMI server maps the addresses in conventional
22916 memory 1:1, i.e.@: the physical (@code{0x00029000} + @code{0x110}) and
22917 linear (@code{0x29110}) addresses are identical.
22919 This command is supported only with some DPMI servers.
22922 @cindex DOS serial data link, remote debugging
22923 In addition to native debugging, the DJGPP port supports remote
22924 debugging via a serial data link. The following commands are specific
22925 to remote serial debugging in the DJGPP port of @value{GDBN}.
22928 @kindex set com1base
22929 @kindex set com1irq
22930 @kindex set com2base
22931 @kindex set com2irq
22932 @kindex set com3base
22933 @kindex set com3irq
22934 @kindex set com4base
22935 @kindex set com4irq
22936 @item set com1base @var{addr}
22937 This command sets the base I/O port address of the @file{COM1} serial
22940 @item set com1irq @var{irq}
22941 This command sets the @dfn{Interrupt Request} (@code{IRQ}) line to use
22942 for the @file{COM1} serial port.
22944 There are similar commands @samp{set com2base}, @samp{set com3irq},
22945 etc.@: for setting the port address and the @code{IRQ} lines for the
22948 @kindex show com1base
22949 @kindex show com1irq
22950 @kindex show com2base
22951 @kindex show com2irq
22952 @kindex show com3base
22953 @kindex show com3irq
22954 @kindex show com4base
22955 @kindex show com4irq
22956 The related commands @samp{show com1base}, @samp{show com1irq} etc.@:
22957 display the current settings of the base address and the @code{IRQ}
22958 lines used by the COM ports.
22961 @kindex info serial
22962 @cindex DOS serial port status
22963 This command prints the status of the 4 DOS serial ports. For each
22964 port, it prints whether it's active or not, its I/O base address and
22965 IRQ number, whether it uses a 16550-style FIFO, its baudrate, and the
22966 counts of various errors encountered so far.
22970 @node Cygwin Native
22971 @subsection Features for Debugging MS Windows PE Executables
22972 @cindex MS Windows debugging
22973 @cindex native Cygwin debugging
22974 @cindex Cygwin-specific commands
22976 @value{GDBN} supports native debugging of MS Windows programs, including
22977 DLLs with and without symbolic debugging information.
22979 @cindex Ctrl-BREAK, MS-Windows
22980 @cindex interrupt debuggee on MS-Windows
22981 MS-Windows programs that call @code{SetConsoleMode} to switch off the
22982 special meaning of the @samp{Ctrl-C} keystroke cannot be interrupted
22983 by typing @kbd{C-c}. For this reason, @value{GDBN} on MS-Windows
22984 supports @kbd{C-@key{BREAK}} as an alternative interrupt key
22985 sequence, which can be used to interrupt the debuggee even if it
22988 There are various additional Cygwin-specific commands, described in
22989 this section. Working with DLLs that have no debugging symbols is
22990 described in @ref{Non-debug DLL Symbols}.
22995 This is a prefix of MS Windows-specific commands which print
22996 information about the target system and important OS structures.
22998 @item info w32 selector
22999 This command displays information returned by
23000 the Win32 API @code{GetThreadSelectorEntry} function.
23001 It takes an optional argument that is evaluated to
23002 a long value to give the information about this given selector.
23003 Without argument, this command displays information
23004 about the six segment registers.
23006 @item info w32 thread-information-block
23007 This command displays thread specific information stored in the
23008 Thread Information Block (readable on the X86 CPU family using @code{$fs}
23009 selector for 32-bit programs and @code{$gs} for 64-bit programs).
23011 @kindex signal-event
23012 @item signal-event @var{id}
23013 This command signals an event with user-provided @var{id}. Used to resume
23014 crashing process when attached to it using MS-Windows JIT debugging (AeDebug).
23016 To use it, create or edit the following keys in
23017 @code{HKLM\SOFTWARE\Microsoft\Windows NT\CurrentVersion\AeDebug} and/or
23018 @code{HKLM\SOFTWARE\Wow6432Node\Microsoft\Windows NT\CurrentVersion\AeDebug}
23019 (for x86_64 versions):
23023 @code{Debugger} (REG_SZ) --- a command to launch the debugger.
23024 Suggested command is: @code{@var{fully-qualified-path-to-gdb.exe} -ex
23025 "attach %ld" -ex "signal-event %ld" -ex "continue"}.
23027 The first @code{%ld} will be replaced by the process ID of the
23028 crashing process, the second @code{%ld} will be replaced by the ID of
23029 the event that blocks the crashing process, waiting for @value{GDBN}
23033 @code{Auto} (REG_SZ) --- either @code{1} or @code{0}. @code{1} will
23034 make the system run debugger specified by the Debugger key
23035 automatically, @code{0} will cause a dialog box with ``OK'' and
23036 ``Cancel'' buttons to appear, which allows the user to either
23037 terminate the crashing process (OK) or debug it (Cancel).
23040 @kindex set cygwin-exceptions
23041 @cindex debugging the Cygwin DLL
23042 @cindex Cygwin DLL, debugging
23043 @item set cygwin-exceptions @var{mode}
23044 If @var{mode} is @code{on}, @value{GDBN} will break on exceptions that
23045 happen inside the Cygwin DLL. If @var{mode} is @code{off},
23046 @value{GDBN} will delay recognition of exceptions, and may ignore some
23047 exceptions which seem to be caused by internal Cygwin DLL
23048 ``bookkeeping''. This option is meant primarily for debugging the
23049 Cygwin DLL itself; the default value is @code{off} to avoid annoying
23050 @value{GDBN} users with false @code{SIGSEGV} signals.
23052 @kindex show cygwin-exceptions
23053 @item show cygwin-exceptions
23054 Displays whether @value{GDBN} will break on exceptions that happen
23055 inside the Cygwin DLL itself.
23057 @kindex set new-console
23058 @item set new-console @var{mode}
23059 If @var{mode} is @code{on} the debuggee will
23060 be started in a new console on next start.
23061 If @var{mode} is @code{off}, the debuggee will
23062 be started in the same console as the debugger.
23064 @kindex show new-console
23065 @item show new-console
23066 Displays whether a new console is used
23067 when the debuggee is started.
23069 @kindex set new-group
23070 @item set new-group @var{mode}
23071 This boolean value controls whether the debuggee should
23072 start a new group or stay in the same group as the debugger.
23073 This affects the way the Windows OS handles
23076 @kindex show new-group
23077 @item show new-group
23078 Displays current value of new-group boolean.
23080 @kindex set debugevents
23081 @item set debugevents
23082 This boolean value adds debug output concerning kernel events related
23083 to the debuggee seen by the debugger. This includes events that
23084 signal thread and process creation and exit, DLL loading and
23085 unloading, console interrupts, and debugging messages produced by the
23086 Windows @code{OutputDebugString} API call.
23088 @kindex set debugexec
23089 @item set debugexec
23090 This boolean value adds debug output concerning execute events
23091 (such as resume thread) seen by the debugger.
23093 @kindex set debugexceptions
23094 @item set debugexceptions
23095 This boolean value adds debug output concerning exceptions in the
23096 debuggee seen by the debugger.
23098 @kindex set debugmemory
23099 @item set debugmemory
23100 This boolean value adds debug output concerning debuggee memory reads
23101 and writes by the debugger.
23105 This boolean values specifies whether the debuggee is called
23106 via a shell or directly (default value is on).
23110 Displays if the debuggee will be started with a shell.
23115 * Non-debug DLL Symbols:: Support for DLLs without debugging symbols
23118 @node Non-debug DLL Symbols
23119 @subsubsection Support for DLLs without Debugging Symbols
23120 @cindex DLLs with no debugging symbols
23121 @cindex Minimal symbols and DLLs
23123 Very often on windows, some of the DLLs that your program relies on do
23124 not include symbolic debugging information (for example,
23125 @file{kernel32.dll}). When @value{GDBN} doesn't recognize any debugging
23126 symbols in a DLL, it relies on the minimal amount of symbolic
23127 information contained in the DLL's export table. This section
23128 describes working with such symbols, known internally to @value{GDBN} as
23129 ``minimal symbols''.
23131 Note that before the debugged program has started execution, no DLLs
23132 will have been loaded. The easiest way around this problem is simply to
23133 start the program --- either by setting a breakpoint or letting the
23134 program run once to completion.
23136 @subsubsection DLL Name Prefixes
23138 In keeping with the naming conventions used by the Microsoft debugging
23139 tools, DLL export symbols are made available with a prefix based on the
23140 DLL name, for instance @code{KERNEL32!CreateFileA}. The plain name is
23141 also entered into the symbol table, so @code{CreateFileA} is often
23142 sufficient. In some cases there will be name clashes within a program
23143 (particularly if the executable itself includes full debugging symbols)
23144 necessitating the use of the fully qualified name when referring to the
23145 contents of the DLL. Use single-quotes around the name to avoid the
23146 exclamation mark (``!'') being interpreted as a language operator.
23148 Note that the internal name of the DLL may be all upper-case, even
23149 though the file name of the DLL is lower-case, or vice-versa. Since
23150 symbols within @value{GDBN} are @emph{case-sensitive} this may cause
23151 some confusion. If in doubt, try the @code{info functions} and
23152 @code{info variables} commands or even @code{maint print msymbols}
23153 (@pxref{Symbols}). Here's an example:
23156 (@value{GDBP}) info function CreateFileA
23157 All functions matching regular expression "CreateFileA":
23159 Non-debugging symbols:
23160 0x77e885f4 CreateFileA
23161 0x77e885f4 KERNEL32!CreateFileA
23165 (@value{GDBP}) info function !
23166 All functions matching regular expression "!":
23168 Non-debugging symbols:
23169 0x6100114c cygwin1!__assert
23170 0x61004034 cygwin1!_dll_crt0@@0
23171 0x61004240 cygwin1!dll_crt0(per_process *)
23175 @subsubsection Working with Minimal Symbols
23177 Symbols extracted from a DLL's export table do not contain very much
23178 type information. All that @value{GDBN} can do is guess whether a symbol
23179 refers to a function or variable depending on the linker section that
23180 contains the symbol. Also note that the actual contents of the memory
23181 contained in a DLL are not available unless the program is running. This
23182 means that you cannot examine the contents of a variable or disassemble
23183 a function within a DLL without a running program.
23185 Variables are generally treated as pointers and dereferenced
23186 automatically. For this reason, it is often necessary to prefix a
23187 variable name with the address-of operator (``&'') and provide explicit
23188 type information in the command. Here's an example of the type of
23192 (@value{GDBP}) print 'cygwin1!__argv'
23193 'cygwin1!__argv' has unknown type; cast it to its declared type
23197 (@value{GDBP}) x 'cygwin1!__argv'
23198 'cygwin1!__argv' has unknown type; cast it to its declared type
23201 And two possible solutions:
23204 (@value{GDBP}) print ((char **)'cygwin1!__argv')[0]
23205 $2 = 0x22fd98 "/cygdrive/c/mydirectory/myprogram"
23209 (@value{GDBP}) x/2x &'cygwin1!__argv'
23210 0x610c0aa8 <cygwin1!__argv>: 0x10021608 0x00000000
23211 (@value{GDBP}) x/x 0x10021608
23212 0x10021608: 0x0022fd98
23213 (@value{GDBP}) x/s 0x0022fd98
23214 0x22fd98: "/cygdrive/c/mydirectory/myprogram"
23217 Setting a break point within a DLL is possible even before the program
23218 starts execution. However, under these circumstances, @value{GDBN} can't
23219 examine the initial instructions of the function in order to skip the
23220 function's frame set-up code. You can work around this by using ``*&''
23221 to set the breakpoint at a raw memory address:
23224 (@value{GDBP}) break *&'python22!PyOS_Readline'
23225 Breakpoint 1 at 0x1e04eff0
23228 The author of these extensions is not entirely convinced that setting a
23229 break point within a shared DLL like @file{kernel32.dll} is completely
23233 @subsection Commands Specific to @sc{gnu} Hurd Systems
23234 @cindex @sc{gnu} Hurd debugging
23236 This subsection describes @value{GDBN} commands specific to the
23237 @sc{gnu} Hurd native debugging.
23242 @kindex set signals@r{, Hurd command}
23243 @kindex set sigs@r{, Hurd command}
23244 This command toggles the state of inferior signal interception by
23245 @value{GDBN}. Mach exceptions, such as breakpoint traps, are not
23246 affected by this command. @code{sigs} is a shorthand alias for
23251 @kindex show signals@r{, Hurd command}
23252 @kindex show sigs@r{, Hurd command}
23253 Show the current state of intercepting inferior's signals.
23255 @item set signal-thread
23256 @itemx set sigthread
23257 @kindex set signal-thread
23258 @kindex set sigthread
23259 This command tells @value{GDBN} which thread is the @code{libc} signal
23260 thread. That thread is run when a signal is delivered to a running
23261 process. @code{set sigthread} is the shorthand alias of @code{set
23264 @item show signal-thread
23265 @itemx show sigthread
23266 @kindex show signal-thread
23267 @kindex show sigthread
23268 These two commands show which thread will run when the inferior is
23269 delivered a signal.
23272 @kindex set stopped@r{, Hurd command}
23273 This commands tells @value{GDBN} that the inferior process is stopped,
23274 as with the @code{SIGSTOP} signal. The stopped process can be
23275 continued by delivering a signal to it.
23278 @kindex show stopped@r{, Hurd command}
23279 This command shows whether @value{GDBN} thinks the debuggee is
23282 @item set exceptions
23283 @kindex set exceptions@r{, Hurd command}
23284 Use this command to turn off trapping of exceptions in the inferior.
23285 When exception trapping is off, neither breakpoints nor
23286 single-stepping will work. To restore the default, set exception
23289 @item show exceptions
23290 @kindex show exceptions@r{, Hurd command}
23291 Show the current state of trapping exceptions in the inferior.
23293 @item set task pause
23294 @kindex set task@r{, Hurd commands}
23295 @cindex task attributes (@sc{gnu} Hurd)
23296 @cindex pause current task (@sc{gnu} Hurd)
23297 This command toggles task suspension when @value{GDBN} has control.
23298 Setting it to on takes effect immediately, and the task is suspended
23299 whenever @value{GDBN} gets control. Setting it to off will take
23300 effect the next time the inferior is continued. If this option is set
23301 to off, you can use @code{set thread default pause on} or @code{set
23302 thread pause on} (see below) to pause individual threads.
23304 @item show task pause
23305 @kindex show task@r{, Hurd commands}
23306 Show the current state of task suspension.
23308 @item set task detach-suspend-count
23309 @cindex task suspend count
23310 @cindex detach from task, @sc{gnu} Hurd
23311 This command sets the suspend count the task will be left with when
23312 @value{GDBN} detaches from it.
23314 @item show task detach-suspend-count
23315 Show the suspend count the task will be left with when detaching.
23317 @item set task exception-port
23318 @itemx set task excp
23319 @cindex task exception port, @sc{gnu} Hurd
23320 This command sets the task exception port to which @value{GDBN} will
23321 forward exceptions. The argument should be the value of the @dfn{send
23322 rights} of the task. @code{set task excp} is a shorthand alias.
23324 @item set noninvasive
23325 @cindex noninvasive task options
23326 This command switches @value{GDBN} to a mode that is the least
23327 invasive as far as interfering with the inferior is concerned. This
23328 is the same as using @code{set task pause}, @code{set exceptions}, and
23329 @code{set signals} to values opposite to the defaults.
23331 @item info send-rights
23332 @itemx info receive-rights
23333 @itemx info port-rights
23334 @itemx info port-sets
23335 @itemx info dead-names
23338 @cindex send rights, @sc{gnu} Hurd
23339 @cindex receive rights, @sc{gnu} Hurd
23340 @cindex port rights, @sc{gnu} Hurd
23341 @cindex port sets, @sc{gnu} Hurd
23342 @cindex dead names, @sc{gnu} Hurd
23343 These commands display information about, respectively, send rights,
23344 receive rights, port rights, port sets, and dead names of a task.
23345 There are also shorthand aliases: @code{info ports} for @code{info
23346 port-rights} and @code{info psets} for @code{info port-sets}.
23348 @item set thread pause
23349 @kindex set thread@r{, Hurd command}
23350 @cindex thread properties, @sc{gnu} Hurd
23351 @cindex pause current thread (@sc{gnu} Hurd)
23352 This command toggles current thread suspension when @value{GDBN} has
23353 control. Setting it to on takes effect immediately, and the current
23354 thread is suspended whenever @value{GDBN} gets control. Setting it to
23355 off will take effect the next time the inferior is continued.
23356 Normally, this command has no effect, since when @value{GDBN} has
23357 control, the whole task is suspended. However, if you used @code{set
23358 task pause off} (see above), this command comes in handy to suspend
23359 only the current thread.
23361 @item show thread pause
23362 @kindex show thread@r{, Hurd command}
23363 This command shows the state of current thread suspension.
23365 @item set thread run
23366 This command sets whether the current thread is allowed to run.
23368 @item show thread run
23369 Show whether the current thread is allowed to run.
23371 @item set thread detach-suspend-count
23372 @cindex thread suspend count, @sc{gnu} Hurd
23373 @cindex detach from thread, @sc{gnu} Hurd
23374 This command sets the suspend count @value{GDBN} will leave on a
23375 thread when detaching. This number is relative to the suspend count
23376 found by @value{GDBN} when it notices the thread; use @code{set thread
23377 takeover-suspend-count} to force it to an absolute value.
23379 @item show thread detach-suspend-count
23380 Show the suspend count @value{GDBN} will leave on the thread when
23383 @item set thread exception-port
23384 @itemx set thread excp
23385 Set the thread exception port to which to forward exceptions. This
23386 overrides the port set by @code{set task exception-port} (see above).
23387 @code{set thread excp} is the shorthand alias.
23389 @item set thread takeover-suspend-count
23390 Normally, @value{GDBN}'s thread suspend counts are relative to the
23391 value @value{GDBN} finds when it notices each thread. This command
23392 changes the suspend counts to be absolute instead.
23394 @item set thread default
23395 @itemx show thread default
23396 @cindex thread default settings, @sc{gnu} Hurd
23397 Each of the above @code{set thread} commands has a @code{set thread
23398 default} counterpart (e.g., @code{set thread default pause}, @code{set
23399 thread default exception-port}, etc.). The @code{thread default}
23400 variety of commands sets the default thread properties for all
23401 threads; you can then change the properties of individual threads with
23402 the non-default commands.
23409 @value{GDBN} provides the following commands specific to the Darwin target:
23412 @item set debug darwin @var{num}
23413 @kindex set debug darwin
23414 When set to a non zero value, enables debugging messages specific to
23415 the Darwin support. Higher values produce more verbose output.
23417 @item show debug darwin
23418 @kindex show debug darwin
23419 Show the current state of Darwin messages.
23421 @item set debug mach-o @var{num}
23422 @kindex set debug mach-o
23423 When set to a non zero value, enables debugging messages while
23424 @value{GDBN} is reading Darwin object files. (@dfn{Mach-O} is the
23425 file format used on Darwin for object and executable files.) Higher
23426 values produce more verbose output. This is a command to diagnose
23427 problems internal to @value{GDBN} and should not be needed in normal
23430 @item show debug mach-o
23431 @kindex show debug mach-o
23432 Show the current state of Mach-O file messages.
23434 @item set mach-exceptions on
23435 @itemx set mach-exceptions off
23436 @kindex set mach-exceptions
23437 On Darwin, faults are first reported as a Mach exception and are then
23438 mapped to a Posix signal. Use this command to turn on trapping of
23439 Mach exceptions in the inferior. This might be sometimes useful to
23440 better understand the cause of a fault. The default is off.
23442 @item show mach-exceptions
23443 @kindex show mach-exceptions
23444 Show the current state of exceptions trapping.
23448 @subsection FreeBSD
23451 When the ABI of a system call is changed in the FreeBSD kernel, this
23452 is implemented by leaving a compatibility system call using the old
23453 ABI at the existing number and allocating a new system call number for
23454 the version using the new ABI. As a convenience, when a system call
23455 is caught by name (@pxref{catch syscall}), compatibility system calls
23458 For example, FreeBSD 12 introduced a new variant of the @code{kevent}
23459 system call and catching the @code{kevent} system call by name catches
23463 (@value{GDBP}) catch syscall kevent
23464 Catchpoint 1 (syscalls 'freebsd11_kevent' [363] 'kevent' [560])
23470 @section Embedded Operating Systems
23472 This section describes configurations involving the debugging of
23473 embedded operating systems that are available for several different
23476 @value{GDBN} includes the ability to debug programs running on
23477 various real-time operating systems.
23479 @node Embedded Processors
23480 @section Embedded Processors
23482 This section goes into details specific to particular embedded
23485 @cindex send command to simulator
23486 Whenever a specific embedded processor has a simulator, @value{GDBN}
23487 allows to send an arbitrary command to the simulator.
23490 @item sim @var{command}
23491 @kindex sim@r{, a command}
23492 Send an arbitrary @var{command} string to the simulator. Consult the
23493 documentation for the specific simulator in use for information about
23494 acceptable commands.
23499 * ARC:: Synopsys ARC
23501 * M68K:: Motorola M68K
23502 * MicroBlaze:: Xilinx MicroBlaze
23503 * MIPS Embedded:: MIPS Embedded
23504 * OpenRISC 1000:: OpenRISC 1000 (or1k)
23505 * PowerPC Embedded:: PowerPC Embedded
23508 * Super-H:: Renesas Super-H
23512 @subsection Synopsys ARC
23513 @cindex Synopsys ARC
23514 @cindex ARC specific commands
23520 @value{GDBN} provides the following ARC-specific commands:
23523 @item set debug arc
23524 @kindex set debug arc
23525 Control the level of ARC specific debug messages. Use 0 for no messages (the
23526 default), 1 for debug messages, and 2 for even more debug messages.
23528 @item show debug arc
23529 @kindex show debug arc
23530 Show the level of ARC specific debugging in operation.
23532 @item maint print arc arc-instruction @var{address}
23533 @kindex maint print arc arc-instruction
23534 Print internal disassembler information about instruction at a given address.
23541 @value{GDBN} provides the following ARM-specific commands:
23544 @item set arm disassembler
23546 This commands selects from a list of disassembly styles. The
23547 @code{"std"} style is the standard style.
23549 @item show arm disassembler
23551 Show the current disassembly style.
23553 @item set arm apcs32
23554 @cindex ARM 32-bit mode
23555 This command toggles ARM operation mode between 32-bit and 26-bit.
23557 @item show arm apcs32
23558 Display the current usage of the ARM 32-bit mode.
23560 @item set arm fpu @var{fputype}
23561 This command sets the ARM floating-point unit (FPU) type. The
23562 argument @var{fputype} can be one of these:
23566 Determine the FPU type by querying the OS ABI.
23568 Software FPU, with mixed-endian doubles on little-endian ARM
23571 GCC-compiled FPA co-processor.
23573 Software FPU with pure-endian doubles.
23579 Show the current type of the FPU.
23582 This command forces @value{GDBN} to use the specified ABI.
23585 Show the currently used ABI.
23587 @item set arm fallback-mode (arm|thumb|auto)
23588 @value{GDBN} uses the symbol table, when available, to determine
23589 whether instructions are ARM or Thumb. This command controls
23590 @value{GDBN}'s default behavior when the symbol table is not
23591 available. The default is @samp{auto}, which causes @value{GDBN} to
23592 use the current execution mode (from the @code{T} bit in the @code{CPSR}
23595 @item show arm fallback-mode
23596 Show the current fallback instruction mode.
23598 @item set arm force-mode (arm|thumb|auto)
23599 This command overrides use of the symbol table to determine whether
23600 instructions are ARM or Thumb. The default is @samp{auto}, which
23601 causes @value{GDBN} to use the symbol table and then the setting
23602 of @samp{set arm fallback-mode}.
23604 @item show arm force-mode
23605 Show the current forced instruction mode.
23607 @item set debug arm
23608 Toggle whether to display ARM-specific debugging messages from the ARM
23609 target support subsystem.
23611 @item show debug arm
23612 Show whether ARM-specific debugging messages are enabled.
23616 @item target sim @r{[}@var{simargs}@r{]} @dots{}
23617 The @value{GDBN} ARM simulator accepts the following optional arguments.
23620 @item --swi-support=@var{type}
23621 Tell the simulator which SWI interfaces to support. The argument
23622 @var{type} may be a comma separated list of the following values.
23623 The default value is @code{all}.
23638 The Motorola m68k configuration includes ColdFire support.
23641 @subsection MicroBlaze
23642 @cindex Xilinx MicroBlaze
23643 @cindex XMD, Xilinx Microprocessor Debugger
23645 The MicroBlaze is a soft-core processor supported on various Xilinx
23646 FPGAs, such as Spartan or Virtex series. Boards with these processors
23647 usually have JTAG ports which connect to a host system running the Xilinx
23648 Embedded Development Kit (EDK) or Software Development Kit (SDK).
23649 This host system is used to download the configuration bitstream to
23650 the target FPGA. The Xilinx Microprocessor Debugger (XMD) program
23651 communicates with the target board using the JTAG interface and
23652 presents a @code{gdbserver} interface to the board. By default
23653 @code{xmd} uses port @code{1234}. (While it is possible to change
23654 this default port, it requires the use of undocumented @code{xmd}
23655 commands. Contact Xilinx support if you need to do this.)
23657 Use these GDB commands to connect to the MicroBlaze target processor.
23660 @item target remote :1234
23661 Use this command to connect to the target if you are running @value{GDBN}
23662 on the same system as @code{xmd}.
23664 @item target remote @var{xmd-host}:1234
23665 Use this command to connect to the target if it is connected to @code{xmd}
23666 running on a different system named @var{xmd-host}.
23669 Use this command to download a program to the MicroBlaze target.
23671 @item set debug microblaze @var{n}
23672 Enable MicroBlaze-specific debugging messages if non-zero.
23674 @item show debug microblaze @var{n}
23675 Show MicroBlaze-specific debugging level.
23678 @node MIPS Embedded
23679 @subsection @acronym{MIPS} Embedded
23682 @value{GDBN} supports these special commands for @acronym{MIPS} targets:
23685 @item set mipsfpu double
23686 @itemx set mipsfpu single
23687 @itemx set mipsfpu none
23688 @itemx set mipsfpu auto
23689 @itemx show mipsfpu
23690 @kindex set mipsfpu
23691 @kindex show mipsfpu
23692 @cindex @acronym{MIPS} remote floating point
23693 @cindex floating point, @acronym{MIPS} remote
23694 If your target board does not support the @acronym{MIPS} floating point
23695 coprocessor, you should use the command @samp{set mipsfpu none} (if you
23696 need this, you may wish to put the command in your @value{GDBN} init
23697 file). This tells @value{GDBN} how to find the return value of
23698 functions which return floating point values. It also allows
23699 @value{GDBN} to avoid saving the floating point registers when calling
23700 functions on the board. If you are using a floating point coprocessor
23701 with only single precision floating point support, as on the @sc{r4650}
23702 processor, use the command @samp{set mipsfpu single}. The default
23703 double precision floating point coprocessor may be selected using
23704 @samp{set mipsfpu double}.
23706 In previous versions the only choices were double precision or no
23707 floating point, so @samp{set mipsfpu on} will select double precision
23708 and @samp{set mipsfpu off} will select no floating point.
23710 As usual, you can inquire about the @code{mipsfpu} variable with
23711 @samp{show mipsfpu}.
23714 @node OpenRISC 1000
23715 @subsection OpenRISC 1000
23716 @cindex OpenRISC 1000
23719 The OpenRISC 1000 provides a free RISC instruction set architecture. It is
23720 mainly provided as a soft-core which can run on Xilinx, Altera and other
23723 @value{GDBN} for OpenRISC supports the below commands when connecting to
23731 Runs the builtin CPU simulator which can run very basic
23732 programs but does not support most hardware functions like MMU.
23733 For more complex use cases the user is advised to run an external
23734 target, and connect using @samp{target remote}.
23736 Example: @code{target sim}
23738 @item set debug or1k
23739 Toggle whether to display OpenRISC-specific debugging messages from the
23740 OpenRISC target support subsystem.
23742 @item show debug or1k
23743 Show whether OpenRISC-specific debugging messages are enabled.
23746 @node PowerPC Embedded
23747 @subsection PowerPC Embedded
23749 @cindex DVC register
23750 @value{GDBN} supports using the DVC (Data Value Compare) register to
23751 implement in hardware simple hardware watchpoint conditions of the form:
23754 (@value{GDBP}) watch @var{ADDRESS|VARIABLE} \
23755 if @var{ADDRESS|VARIABLE} == @var{CONSTANT EXPRESSION}
23758 The DVC register will be automatically used when @value{GDBN} detects
23759 such pattern in a condition expression, and the created watchpoint uses one
23760 debug register (either the @code{exact-watchpoints} option is on and the
23761 variable is scalar, or the variable has a length of one byte). This feature
23762 is available in native @value{GDBN} running on a Linux kernel version 2.6.34
23765 When running on PowerPC embedded processors, @value{GDBN} automatically uses
23766 ranged hardware watchpoints, unless the @code{exact-watchpoints} option is on,
23767 in which case watchpoints using only one debug register are created when
23768 watching variables of scalar types.
23770 You can create an artificial array to watch an arbitrary memory
23771 region using one of the following commands (@pxref{Expressions}):
23774 (@value{GDBP}) watch *((char *) @var{address})@@@var{length}
23775 (@value{GDBP}) watch @{char[@var{length}]@} @var{address}
23778 PowerPC embedded processors support masked watchpoints. See the discussion
23779 about the @code{mask} argument in @ref{Set Watchpoints}.
23781 @cindex ranged breakpoint
23782 PowerPC embedded processors support hardware accelerated
23783 @dfn{ranged breakpoints}. A ranged breakpoint stops execution of
23784 the inferior whenever it executes an instruction at any address within
23785 the range it specifies. To set a ranged breakpoint in @value{GDBN},
23786 use the @code{break-range} command.
23788 @value{GDBN} provides the following PowerPC-specific commands:
23791 @kindex break-range
23792 @item break-range @var{start-location}, @var{end-location}
23793 Set a breakpoint for an address range given by
23794 @var{start-location} and @var{end-location}, which can specify a function name,
23795 a line number, an offset of lines from the current line or from the start
23796 location, or an address of an instruction (see @ref{Specify Location},
23797 for a list of all the possible ways to specify a @var{location}.)
23798 The breakpoint will stop execution of the inferior whenever it
23799 executes an instruction at any address within the specified range,
23800 (including @var{start-location} and @var{end-location}.)
23802 @kindex set powerpc
23803 @item set powerpc soft-float
23804 @itemx show powerpc soft-float
23805 Force @value{GDBN} to use (or not use) a software floating point calling
23806 convention. By default, @value{GDBN} selects the calling convention based
23807 on the selected architecture and the provided executable file.
23809 @item set powerpc vector-abi
23810 @itemx show powerpc vector-abi
23811 Force @value{GDBN} to use the specified calling convention for vector
23812 arguments and return values. The valid options are @samp{auto};
23813 @samp{generic}, to avoid vector registers even if they are present;
23814 @samp{altivec}, to use AltiVec registers; and @samp{spe} to use SPE
23815 registers. By default, @value{GDBN} selects the calling convention
23816 based on the selected architecture and the provided executable file.
23818 @item set powerpc exact-watchpoints
23819 @itemx show powerpc exact-watchpoints
23820 Allow @value{GDBN} to use only one debug register when watching a variable
23821 of scalar type, thus assuming that the variable is accessed through the
23822 address of its first byte.
23827 @subsection Atmel AVR
23830 When configured for debugging the Atmel AVR, @value{GDBN} supports the
23831 following AVR-specific commands:
23834 @item info io_registers
23835 @kindex info io_registers@r{, AVR}
23836 @cindex I/O registers (Atmel AVR)
23837 This command displays information about the AVR I/O registers. For
23838 each register, @value{GDBN} prints its number and value.
23845 When configured for debugging CRIS, @value{GDBN} provides the
23846 following CRIS-specific commands:
23849 @item set cris-version @var{ver}
23850 @cindex CRIS version
23851 Set the current CRIS version to @var{ver}, either @samp{10} or @samp{32}.
23852 The CRIS version affects register names and sizes. This command is useful in
23853 case autodetection of the CRIS version fails.
23855 @item show cris-version
23856 Show the current CRIS version.
23858 @item set cris-dwarf2-cfi
23859 @cindex DWARF-2 CFI and CRIS
23860 Set the usage of DWARF-2 CFI for CRIS debugging. The default is @samp{on}.
23861 Change to @samp{off} when using @code{gcc-cris} whose version is below
23864 @item show cris-dwarf2-cfi
23865 Show the current state of using DWARF-2 CFI.
23867 @item set cris-mode @var{mode}
23869 Set the current CRIS mode to @var{mode}. It should only be changed when
23870 debugging in guru mode, in which case it should be set to
23871 @samp{guru} (the default is @samp{normal}).
23873 @item show cris-mode
23874 Show the current CRIS mode.
23878 @subsection Renesas Super-H
23881 For the Renesas Super-H processor, @value{GDBN} provides these
23885 @item set sh calling-convention @var{convention}
23886 @kindex set sh calling-convention
23887 Set the calling-convention used when calling functions from @value{GDBN}.
23888 Allowed values are @samp{gcc}, which is the default setting, and @samp{renesas}.
23889 With the @samp{gcc} setting, functions are called using the @value{NGCC} calling
23890 convention. If the DWARF-2 information of the called function specifies
23891 that the function follows the Renesas calling convention, the function
23892 is called using the Renesas calling convention. If the calling convention
23893 is set to @samp{renesas}, the Renesas calling convention is always used,
23894 regardless of the DWARF-2 information. This can be used to override the
23895 default of @samp{gcc} if debug information is missing, or the compiler
23896 does not emit the DWARF-2 calling convention entry for a function.
23898 @item show sh calling-convention
23899 @kindex show sh calling-convention
23900 Show the current calling convention setting.
23905 @node Architectures
23906 @section Architectures
23908 This section describes characteristics of architectures that affect
23909 all uses of @value{GDBN} with the architecture, both native and cross.
23916 * HPPA:: HP PA architecture
23917 * SPU:: Cell Broadband Engine SPU architecture
23925 @subsection AArch64
23926 @cindex AArch64 support
23928 When @value{GDBN} is debugging the AArch64 architecture, it provides the
23929 following special commands:
23932 @item set debug aarch64
23933 @kindex set debug aarch64
23934 This command determines whether AArch64 architecture-specific debugging
23935 messages are to be displayed.
23937 @item show debug aarch64
23938 Show whether AArch64 debugging messages are displayed.
23942 @subsubsection AArch64 SVE.
23943 @cindex AArch64 SVE.
23945 When @value{GDBN} is debugging the AArch64 architecture, if the Scalable Vector
23946 Extension (SVE) is present, then @value{GDBN} will provide the vector registers
23947 @code{$z0} through @code{$z31}, vector predicate registers @code{$p0} through
23948 @code{$p15}, and the @code{$ffr} register. In addition, the pseudo register
23949 @code{$vg} will be provided. This is the vector granule for the current thread
23950 and represents the number of 64-bit chunks in an SVE @code{z} register.
23952 If the vector length changes, then the @code{$vg} register will be updated,
23953 but the lengths of the @code{z} and @code{p} registers will not change. This
23954 is a known limitation of @value{GDBN} and does not affect the execution of the
23959 @subsection x86 Architecture-specific Issues
23962 @item set struct-convention @var{mode}
23963 @kindex set struct-convention
23964 @cindex struct return convention
23965 @cindex struct/union returned in registers
23966 Set the convention used by the inferior to return @code{struct}s and
23967 @code{union}s from functions to @var{mode}. Possible values of
23968 @var{mode} are @code{"pcc"}, @code{"reg"}, and @code{"default"} (the
23969 default). @code{"default"} or @code{"pcc"} means that @code{struct}s
23970 are returned on the stack, while @code{"reg"} means that a
23971 @code{struct} or a @code{union} whose size is 1, 2, 4, or 8 bytes will
23972 be returned in a register.
23974 @item show struct-convention
23975 @kindex show struct-convention
23976 Show the current setting of the convention to return @code{struct}s
23981 @subsubsection Intel @dfn{Memory Protection Extensions} (MPX).
23982 @cindex Intel Memory Protection Extensions (MPX).
23984 Memory Protection Extension (MPX) adds the bound registers @samp{BND0}
23985 @footnote{The register named with capital letters represent the architecture
23986 registers.} through @samp{BND3}. Bound registers store a pair of 64-bit values
23987 which are the lower bound and upper bound. Bounds are effective addresses or
23988 memory locations. The upper bounds are architecturally represented in 1's
23989 complement form. A bound having lower bound = 0, and upper bound = 0
23990 (1's complement of all bits set) will allow access to the entire address space.
23992 @samp{BND0} through @samp{BND3} are represented in @value{GDBN} as @samp{bnd0raw}
23993 through @samp{bnd3raw}. Pseudo registers @samp{bnd0} through @samp{bnd3}
23994 display the upper bound performing the complement of one operation on the
23995 upper bound value, i.e.@ when upper bound in @samp{bnd0raw} is 0 in the
23996 @value{GDBN} @samp{bnd0} it will be @code{0xfff@dots{}}. In this sense it
23997 can also be noted that the upper bounds are inclusive.
23999 As an example, assume that the register BND0 holds bounds for a pointer having
24000 access allowed for the range between 0x32 and 0x71. The values present on
24001 bnd0raw and bnd registers are presented as follows:
24004 bnd0raw = @{0x32, 0xffffffff8e@}
24005 bnd0 = @{lbound = 0x32, ubound = 0x71@} : size 64
24008 This way the raw value can be accessed via bnd0raw@dots{}bnd3raw. Any
24009 change on bnd0@dots{}bnd3 or bnd0raw@dots{}bnd3raw is reflect on its
24010 counterpart. When the bnd0@dots{}bnd3 registers are displayed via
24011 Python, the display includes the memory size, in bits, accessible to
24014 Bounds can also be stored in bounds tables, which are stored in
24015 application memory. These tables store bounds for pointers by specifying
24016 the bounds pointer's value along with its bounds. Evaluating and changing
24017 bounds located in bound tables is therefore interesting while investigating
24018 bugs on MPX context. @value{GDBN} provides commands for this purpose:
24021 @item show mpx bound @var{pointer}
24022 @kindex show mpx bound
24023 Display bounds of the given @var{pointer}.
24025 @item set mpx bound @var{pointer}, @var{lbound}, @var{ubound}
24026 @kindex set mpx bound
24027 Set the bounds of a pointer in the bound table.
24028 This command takes three parameters: @var{pointer} is the pointers
24029 whose bounds are to be changed, @var{lbound} and @var{ubound} are new values
24030 for lower and upper bounds respectively.
24033 When you call an inferior function on an Intel MPX enabled program,
24034 GDB sets the inferior's bound registers to the init (disabled) state
24035 before calling the function. As a consequence, bounds checks for the
24036 pointer arguments passed to the function will always pass.
24038 This is necessary because when you call an inferior function, the
24039 program is usually in the middle of the execution of other function.
24040 Since at that point bound registers are in an arbitrary state, not
24041 clearing them would lead to random bound violations in the called
24044 You can still examine the influence of the bound registers on the
24045 execution of the called function by stopping the execution of the
24046 called function at its prologue, setting bound registers, and
24047 continuing the execution. For example:
24051 Breakpoint 2 at 0x4009de: file i386-mpx-call.c, line 47.
24052 $ print upper (a, b, c, d, 1)
24053 Breakpoint 2, upper (a=0x0, b=0x6e0000005b, c=0x0, d=0x0, len=48)....
24055 @{lbound = 0x0, ubound = ffffffff@} : size -1
24058 At this last step the value of bnd0 can be changed for investigation of bound
24059 violations caused along the execution of the call. In order to know how to
24060 set the bound registers or bound table for the call consult the ABI.
24065 See the following section.
24068 @subsection @acronym{MIPS}
24070 @cindex stack on Alpha
24071 @cindex stack on @acronym{MIPS}
24072 @cindex Alpha stack
24073 @cindex @acronym{MIPS} stack
24074 Alpha- and @acronym{MIPS}-based computers use an unusual stack frame, which
24075 sometimes requires @value{GDBN} to search backward in the object code to
24076 find the beginning of a function.
24078 @cindex response time, @acronym{MIPS} debugging
24079 To improve response time (especially for embedded applications, where
24080 @value{GDBN} may be restricted to a slow serial line for this search)
24081 you may want to limit the size of this search, using one of these
24085 @cindex @code{heuristic-fence-post} (Alpha, @acronym{MIPS})
24086 @item set heuristic-fence-post @var{limit}
24087 Restrict @value{GDBN} to examining at most @var{limit} bytes in its
24088 search for the beginning of a function. A value of @var{0} (the
24089 default) means there is no limit. However, except for @var{0}, the
24090 larger the limit the more bytes @code{heuristic-fence-post} must search
24091 and therefore the longer it takes to run. You should only need to use
24092 this command when debugging a stripped executable.
24094 @item show heuristic-fence-post
24095 Display the current limit.
24099 These commands are available @emph{only} when @value{GDBN} is configured
24100 for debugging programs on Alpha or @acronym{MIPS} processors.
24102 Several @acronym{MIPS}-specific commands are available when debugging @acronym{MIPS}
24106 @item set mips abi @var{arg}
24107 @kindex set mips abi
24108 @cindex set ABI for @acronym{MIPS}
24109 Tell @value{GDBN} which @acronym{MIPS} ABI is used by the inferior. Possible
24110 values of @var{arg} are:
24114 The default ABI associated with the current binary (this is the
24124 @item show mips abi
24125 @kindex show mips abi
24126 Show the @acronym{MIPS} ABI used by @value{GDBN} to debug the inferior.
24128 @item set mips compression @var{arg}
24129 @kindex set mips compression
24130 @cindex code compression, @acronym{MIPS}
24131 Tell @value{GDBN} which @acronym{MIPS} compressed
24132 @acronym{ISA, Instruction Set Architecture} encoding is used by the
24133 inferior. @value{GDBN} uses this for code disassembly and other
24134 internal interpretation purposes. This setting is only referred to
24135 when no executable has been associated with the debugging session or
24136 the executable does not provide information about the encoding it uses.
24137 Otherwise this setting is automatically updated from information
24138 provided by the executable.
24140 Possible values of @var{arg} are @samp{mips16} and @samp{micromips}.
24141 The default compressed @acronym{ISA} encoding is @samp{mips16}, as
24142 executables containing @acronym{MIPS16} code frequently are not
24143 identified as such.
24145 This setting is ``sticky''; that is, it retains its value across
24146 debugging sessions until reset either explicitly with this command or
24147 implicitly from an executable.
24149 The compiler and/or assembler typically add symbol table annotations to
24150 identify functions compiled for the @acronym{MIPS16} or
24151 @acronym{microMIPS} @acronym{ISA}s. If these function-scope annotations
24152 are present, @value{GDBN} uses them in preference to the global
24153 compressed @acronym{ISA} encoding setting.
24155 @item show mips compression
24156 @kindex show mips compression
24157 Show the @acronym{MIPS} compressed @acronym{ISA} encoding used by
24158 @value{GDBN} to debug the inferior.
24161 @itemx show mipsfpu
24162 @xref{MIPS Embedded, set mipsfpu}.
24164 @item set mips mask-address @var{arg}
24165 @kindex set mips mask-address
24166 @cindex @acronym{MIPS} addresses, masking
24167 This command determines whether the most-significant 32 bits of 64-bit
24168 @acronym{MIPS} addresses are masked off. The argument @var{arg} can be
24169 @samp{on}, @samp{off}, or @samp{auto}. The latter is the default
24170 setting, which lets @value{GDBN} determine the correct value.
24172 @item show mips mask-address
24173 @kindex show mips mask-address
24174 Show whether the upper 32 bits of @acronym{MIPS} addresses are masked off or
24177 @item set remote-mips64-transfers-32bit-regs
24178 @kindex set remote-mips64-transfers-32bit-regs
24179 This command controls compatibility with 64-bit @acronym{MIPS} targets that
24180 transfer data in 32-bit quantities. If you have an old @acronym{MIPS} 64 target
24181 that transfers 32 bits for some registers, like @sc{sr} and @sc{fsr},
24182 and 64 bits for other registers, set this option to @samp{on}.
24184 @item show remote-mips64-transfers-32bit-regs
24185 @kindex show remote-mips64-transfers-32bit-regs
24186 Show the current setting of compatibility with older @acronym{MIPS} 64 targets.
24188 @item set debug mips
24189 @kindex set debug mips
24190 This command turns on and off debugging messages for the @acronym{MIPS}-specific
24191 target code in @value{GDBN}.
24193 @item show debug mips
24194 @kindex show debug mips
24195 Show the current setting of @acronym{MIPS} debugging messages.
24201 @cindex HPPA support
24203 When @value{GDBN} is debugging the HP PA architecture, it provides the
24204 following special commands:
24207 @item set debug hppa
24208 @kindex set debug hppa
24209 This command determines whether HPPA architecture-specific debugging
24210 messages are to be displayed.
24212 @item show debug hppa
24213 Show whether HPPA debugging messages are displayed.
24215 @item maint print unwind @var{address}
24216 @kindex maint print unwind@r{, HPPA}
24217 This command displays the contents of the unwind table entry at the
24218 given @var{address}.
24224 @subsection Cell Broadband Engine SPU architecture
24225 @cindex Cell Broadband Engine
24228 When @value{GDBN} is debugging the Cell Broadband Engine SPU architecture,
24229 it provides the following special commands:
24232 @item info spu event
24234 Display SPU event facility status. Shows current event mask
24235 and pending event status.
24237 @item info spu signal
24238 Display SPU signal notification facility status. Shows pending
24239 signal-control word and signal notification mode of both signal
24240 notification channels.
24242 @item info spu mailbox
24243 Display SPU mailbox facility status. Shows all pending entries,
24244 in order of processing, in each of the SPU Write Outbound,
24245 SPU Write Outbound Interrupt, and SPU Read Inbound mailboxes.
24248 Display MFC DMA status. Shows all pending commands in the MFC
24249 DMA queue. For each entry, opcode, tag, class IDs, effective
24250 and local store addresses and transfer size are shown.
24252 @item info spu proxydma
24253 Display MFC Proxy-DMA status. Shows all pending commands in the MFC
24254 Proxy-DMA queue. For each entry, opcode, tag, class IDs, effective
24255 and local store addresses and transfer size are shown.
24259 When @value{GDBN} is debugging a combined PowerPC/SPU application
24260 on the Cell Broadband Engine, it provides in addition the following
24264 @item set spu stop-on-load @var{arg}
24266 Set whether to stop for new SPE threads. When set to @code{on}, @value{GDBN}
24267 will give control to the user when a new SPE thread enters its @code{main}
24268 function. The default is @code{off}.
24270 @item show spu stop-on-load
24272 Show whether to stop for new SPE threads.
24274 @item set spu auto-flush-cache @var{arg}
24275 Set whether to automatically flush the software-managed cache. When set to
24276 @code{on}, @value{GDBN} will automatically cause the SPE software-managed
24277 cache to be flushed whenever SPE execution stops. This provides a consistent
24278 view of PowerPC memory that is accessed via the cache. If an application
24279 does not use the software-managed cache, this option has no effect.
24281 @item show spu auto-flush-cache
24282 Show whether to automatically flush the software-managed cache.
24287 @subsection PowerPC
24288 @cindex PowerPC architecture
24290 When @value{GDBN} is debugging the PowerPC architecture, it provides a set of
24291 pseudo-registers to enable inspection of 128-bit wide Decimal Floating Point
24292 numbers stored in the floating point registers. These values must be stored
24293 in two consecutive registers, always starting at an even register like
24294 @code{f0} or @code{f2}.
24296 The pseudo-registers go from @code{$dl0} through @code{$dl15}, and are formed
24297 by joining the even/odd register pairs @code{f0} and @code{f1} for @code{$dl0},
24298 @code{f2} and @code{f3} for @code{$dl1} and so on.
24300 For POWER7 processors, @value{GDBN} provides a set of pseudo-registers, the 64-bit
24301 wide Extended Floating Point Registers (@samp{f32} through @samp{f63}).
24304 @subsection Nios II
24305 @cindex Nios II architecture
24307 When @value{GDBN} is debugging the Nios II architecture,
24308 it provides the following special commands:
24312 @item set debug nios2
24313 @kindex set debug nios2
24314 This command turns on and off debugging messages for the Nios II
24315 target code in @value{GDBN}.
24317 @item show debug nios2
24318 @kindex show debug nios2
24319 Show the current setting of Nios II debugging messages.
24323 @subsection Sparc64
24324 @cindex Sparc64 support
24325 @cindex Application Data Integrity
24326 @subsubsection ADI Support
24328 The M7 processor supports an Application Data Integrity (ADI) feature that
24329 detects invalid data accesses. When software allocates memory and enables
24330 ADI on the allocated memory, it chooses a 4-bit version number, sets the
24331 version in the upper 4 bits of the 64-bit pointer to that data, and stores
24332 the 4-bit version in every cacheline of that data. Hardware saves the latter
24333 in spare bits in the cache and memory hierarchy. On each load and store,
24334 the processor compares the upper 4 VA (virtual address) bits to the
24335 cacheline's version. If there is a mismatch, the processor generates a
24336 version mismatch trap which can be either precise or disrupting. The trap
24337 is an error condition which the kernel delivers to the process as a SIGSEGV
24340 Note that only 64-bit applications can use ADI and need to be built with
24343 Values of the ADI version tags, which are in granularity of a
24344 cacheline (64 bytes), can be viewed or modified.
24348 @kindex adi examine
24349 @item adi (examine | x) [ / @var{n} ] @var{addr}
24351 The @code{adi examine} command displays the value of one ADI version tag per
24354 @var{n} is a decimal integer specifying the number in bytes; the default
24355 is 1. It specifies how much ADI version information, at the ratio of 1:ADI
24356 block size, to display.
24358 @var{addr} is the address in user address space where you want @value{GDBN}
24359 to begin displaying the ADI version tags.
24361 Below is an example of displaying ADI versions of variable "shmaddr".
24364 (@value{GDBP}) adi x/100 shmaddr
24365 0xfff800010002c000: 0 0
24369 @item adi (assign | a) [ / @var{n} ] @var{addr} = @var{tag}
24371 The @code{adi assign} command is used to assign new ADI version tag
24374 @var{n} is a decimal integer specifying the number in bytes;
24375 the default is 1. It specifies how much ADI version information, at the
24376 ratio of 1:ADI block size, to modify.
24378 @var{addr} is the address in user address space where you want @value{GDBN}
24379 to begin modifying the ADI version tags.
24381 @var{tag} is the new ADI version tag.
24383 For example, do the following to modify then verify ADI versions of
24384 variable "shmaddr":
24387 (@value{GDBP}) adi a/100 shmaddr = 7
24388 (@value{GDBP}) adi x/100 shmaddr
24389 0xfff800010002c000: 7 7
24396 @cindex S12Z support
24398 When @value{GDBN} is debugging the S12Z architecture,
24399 it provides the following special command:
24402 @item maint info bdccsr
24403 @kindex maint info bdccsr@r{, S12Z}
24404 This command displays the current value of the microprocessor's
24409 @node Controlling GDB
24410 @chapter Controlling @value{GDBN}
24412 You can alter the way @value{GDBN} interacts with you by using the
24413 @code{set} command. For commands controlling how @value{GDBN} displays
24414 data, see @ref{Print Settings, ,Print Settings}. Other settings are
24419 * Editing:: Command editing
24420 * Command History:: Command history
24421 * Screen Size:: Screen size
24422 * Output Styling:: Output styling
24423 * Numbers:: Numbers
24424 * ABI:: Configuring the current ABI
24425 * Auto-loading:: Automatically loading associated files
24426 * Messages/Warnings:: Optional warnings and messages
24427 * Debugging Output:: Optional messages about internal happenings
24428 * Other Misc Settings:: Other Miscellaneous Settings
24436 @value{GDBN} indicates its readiness to read a command by printing a string
24437 called the @dfn{prompt}. This string is normally @samp{(@value{GDBP})}. You
24438 can change the prompt string with the @code{set prompt} command. For
24439 instance, when debugging @value{GDBN} with @value{GDBN}, it is useful to change
24440 the prompt in one of the @value{GDBN} sessions so that you can always tell
24441 which one you are talking to.
24443 @emph{Note:} @code{set prompt} does not add a space for you after the
24444 prompt you set. This allows you to set a prompt which ends in a space
24445 or a prompt that does not.
24449 @item set prompt @var{newprompt}
24450 Directs @value{GDBN} to use @var{newprompt} as its prompt string henceforth.
24452 @kindex show prompt
24454 Prints a line of the form: @samp{Gdb's prompt is: @var{your-prompt}}
24457 Versions of @value{GDBN} that ship with Python scripting enabled have
24458 prompt extensions. The commands for interacting with these extensions
24462 @kindex set extended-prompt
24463 @item set extended-prompt @var{prompt}
24464 Set an extended prompt that allows for substitutions.
24465 @xref{gdb.prompt}, for a list of escape sequences that can be used for
24466 substitution. Any escape sequences specified as part of the prompt
24467 string are replaced with the corresponding strings each time the prompt
24473 set extended-prompt Current working directory: \w (gdb)
24476 Note that when an extended-prompt is set, it takes control of the
24477 @var{prompt_hook} hook. @xref{prompt_hook}, for further information.
24479 @kindex show extended-prompt
24480 @item show extended-prompt
24481 Prints the extended prompt. Any escape sequences specified as part of
24482 the prompt string with @code{set extended-prompt}, are replaced with the
24483 corresponding strings each time the prompt is displayed.
24487 @section Command Editing
24489 @cindex command line editing
24491 @value{GDBN} reads its input commands via the @dfn{Readline} interface. This
24492 @sc{gnu} library provides consistent behavior for programs which provide a
24493 command line interface to the user. Advantages are @sc{gnu} Emacs-style
24494 or @dfn{vi}-style inline editing of commands, @code{csh}-like history
24495 substitution, and a storage and recall of command history across
24496 debugging sessions.
24498 You may control the behavior of command line editing in @value{GDBN} with the
24499 command @code{set}.
24502 @kindex set editing
24505 @itemx set editing on
24506 Enable command line editing (enabled by default).
24508 @item set editing off
24509 Disable command line editing.
24511 @kindex show editing
24513 Show whether command line editing is enabled.
24516 @ifset SYSTEM_READLINE
24517 @xref{Command Line Editing, , , rluserman, GNU Readline Library},
24519 @ifclear SYSTEM_READLINE
24520 @xref{Command Line Editing},
24522 for more details about the Readline
24523 interface. Users unfamiliar with @sc{gnu} Emacs or @code{vi} are
24524 encouraged to read that chapter.
24526 @node Command History
24527 @section Command History
24528 @cindex command history
24530 @value{GDBN} can keep track of the commands you type during your
24531 debugging sessions, so that you can be certain of precisely what
24532 happened. Use these commands to manage the @value{GDBN} command
24535 @value{GDBN} uses the @sc{gnu} History library, a part of the Readline
24536 package, to provide the history facility.
24537 @ifset SYSTEM_READLINE
24538 @xref{Using History Interactively, , , history, GNU History Library},
24540 @ifclear SYSTEM_READLINE
24541 @xref{Using History Interactively},
24543 for the detailed description of the History library.
24545 To issue a command to @value{GDBN} without affecting certain aspects of
24546 the state which is seen by users, prefix it with @samp{server }
24547 (@pxref{Server Prefix}). This
24548 means that this command will not affect the command history, nor will it
24549 affect @value{GDBN}'s notion of which command to repeat if @key{RET} is
24550 pressed on a line by itself.
24552 @cindex @code{server}, command prefix
24553 The server prefix does not affect the recording of values into the value
24554 history; to print a value without recording it into the value history,
24555 use the @code{output} command instead of the @code{print} command.
24557 Here is the description of @value{GDBN} commands related to command
24561 @cindex history substitution
24562 @cindex history file
24563 @kindex set history filename
24564 @cindex @env{GDBHISTFILE}, environment variable
24565 @item set history filename @var{fname}
24566 Set the name of the @value{GDBN} command history file to @var{fname}.
24567 This is the file where @value{GDBN} reads an initial command history
24568 list, and where it writes the command history from this session when it
24569 exits. You can access this list through history expansion or through
24570 the history command editing characters listed below. This file defaults
24571 to the value of the environment variable @code{GDBHISTFILE}, or to
24572 @file{./.gdb_history} (@file{./_gdb_history} on MS-DOS) if this variable
24575 @cindex save command history
24576 @kindex set history save
24577 @item set history save
24578 @itemx set history save on
24579 Record command history in a file, whose name may be specified with the
24580 @code{set history filename} command. By default, this option is disabled.
24582 @item set history save off
24583 Stop recording command history in a file.
24585 @cindex history size
24586 @kindex set history size
24587 @cindex @env{GDBHISTSIZE}, environment variable
24588 @item set history size @var{size}
24589 @itemx set history size unlimited
24590 Set the number of commands which @value{GDBN} keeps in its history list.
24591 This defaults to the value of the environment variable @env{GDBHISTSIZE}, or
24592 to 256 if this variable is not set. Non-numeric values of @env{GDBHISTSIZE}
24593 are ignored. If @var{size} is @code{unlimited} or if @env{GDBHISTSIZE} is
24594 either a negative number or the empty string, then the number of commands
24595 @value{GDBN} keeps in the history list is unlimited.
24597 @cindex remove duplicate history
24598 @kindex set history remove-duplicates
24599 @item set history remove-duplicates @var{count}
24600 @itemx set history remove-duplicates unlimited
24601 Control the removal of duplicate history entries in the command history list.
24602 If @var{count} is non-zero, @value{GDBN} will look back at the last @var{count}
24603 history entries and remove the first entry that is a duplicate of the current
24604 entry being added to the command history list. If @var{count} is
24605 @code{unlimited} then this lookbehind is unbounded. If @var{count} is 0, then
24606 removal of duplicate history entries is disabled.
24608 Only history entries added during the current session are considered for
24609 removal. This option is set to 0 by default.
24613 History expansion assigns special meaning to the character @kbd{!}.
24614 @ifset SYSTEM_READLINE
24615 @xref{Event Designators, , , history, GNU History Library},
24617 @ifclear SYSTEM_READLINE
24618 @xref{Event Designators},
24622 @cindex history expansion, turn on/off
24623 Since @kbd{!} is also the logical not operator in C, history expansion
24624 is off by default. If you decide to enable history expansion with the
24625 @code{set history expansion on} command, you may sometimes need to
24626 follow @kbd{!} (when it is used as logical not, in an expression) with
24627 a space or a tab to prevent it from being expanded. The readline
24628 history facilities do not attempt substitution on the strings
24629 @kbd{!=} and @kbd{!(}, even when history expansion is enabled.
24631 The commands to control history expansion are:
24634 @item set history expansion on
24635 @itemx set history expansion
24636 @kindex set history expansion
24637 Enable history expansion. History expansion is off by default.
24639 @item set history expansion off
24640 Disable history expansion.
24643 @kindex show history
24645 @itemx show history filename
24646 @itemx show history save
24647 @itemx show history size
24648 @itemx show history expansion
24649 These commands display the state of the @value{GDBN} history parameters.
24650 @code{show history} by itself displays all four states.
24655 @kindex show commands
24656 @cindex show last commands
24657 @cindex display command history
24658 @item show commands
24659 Display the last ten commands in the command history.
24661 @item show commands @var{n}
24662 Print ten commands centered on command number @var{n}.
24664 @item show commands +
24665 Print ten commands just after the commands last printed.
24669 @section Screen Size
24670 @cindex size of screen
24671 @cindex screen size
24674 @cindex pauses in output
24676 Certain commands to @value{GDBN} may produce large amounts of
24677 information output to the screen. To help you read all of it,
24678 @value{GDBN} pauses and asks you for input at the end of each page of
24679 output. Type @key{RET} when you want to see one more page of output,
24680 @kbd{q} to discard the remaining output, or @kbd{c} to continue
24681 without paging for the rest of the current command. Also, the screen
24682 width setting determines when to wrap lines of output. Depending on
24683 what is being printed, @value{GDBN} tries to break the line at a
24684 readable place, rather than simply letting it overflow onto the
24687 Normally @value{GDBN} knows the size of the screen from the terminal
24688 driver software. For example, on Unix @value{GDBN} uses the termcap data base
24689 together with the value of the @code{TERM} environment variable and the
24690 @code{stty rows} and @code{stty cols} settings. If this is not correct,
24691 you can override it with the @code{set height} and @code{set
24698 @kindex show height
24699 @item set height @var{lpp}
24700 @itemx set height unlimited
24702 @itemx set width @var{cpl}
24703 @itemx set width unlimited
24705 These @code{set} commands specify a screen height of @var{lpp} lines and
24706 a screen width of @var{cpl} characters. The associated @code{show}
24707 commands display the current settings.
24709 If you specify a height of either @code{unlimited} or zero lines,
24710 @value{GDBN} does not pause during output no matter how long the
24711 output is. This is useful if output is to a file or to an editor
24714 Likewise, you can specify @samp{set width unlimited} or @samp{set
24715 width 0} to prevent @value{GDBN} from wrapping its output.
24717 @item set pagination on
24718 @itemx set pagination off
24719 @kindex set pagination
24720 Turn the output pagination on or off; the default is on. Turning
24721 pagination off is the alternative to @code{set height unlimited}. Note that
24722 running @value{GDBN} with the @option{--batch} option (@pxref{Mode
24723 Options, -batch}) also automatically disables pagination.
24725 @item show pagination
24726 @kindex show pagination
24727 Show the current pagination mode.
24730 @node Output Styling
24731 @section Output Styling
24737 @value{GDBN} can style its output on a capable terminal. This is
24738 enabled by default on most systems, but disabled by default when in
24739 batch mode (@pxref{Mode Options}). Various style settings are available;
24740 and styles can also be disabled entirely.
24743 @item set style enabled @samp{on|off}
24744 Enable or disable all styling. The default is host-dependent, with
24745 most hosts defaulting to @samp{on}.
24747 @item show style enabled
24748 Show the current state of styling.
24750 @item set style sources @samp{on|off}
24751 Enable or disable source code styling. This affects whether source
24752 code, such as the output of the @code{list} command, is styled. Note
24753 that source styling only works if styling in general is enabled, and
24754 if @value{GDBN} was linked with the GNU Source Highlight library. The
24755 default is @samp{on}.
24757 @item show style sources
24758 Show the current state of source code styling.
24761 Subcommands of @code{set style} control specific forms of styling.
24762 These subcommands all follow the same pattern: each style-able object
24763 can be styled with a foreground color, a background color, and an
24766 For example, the style of file names can be controlled using the
24767 @code{set style filename} group of commands:
24770 @item set style filename background @var{color}
24771 Set the background to @var{color}. Valid colors are @samp{none}
24772 (meaning the terminal's default color), @samp{black}, @samp{red},
24773 @samp{green}, @samp{yellow}, @samp{blue}, @samp{magenta}, @samp{cyan},
24776 @item set style filename foreground @var{color}
24777 Set the foreground to @var{color}. Valid colors are @samp{none}
24778 (meaning the terminal's default color), @samp{black}, @samp{red},
24779 @samp{green}, @samp{yellow}, @samp{blue}, @samp{magenta}, @samp{cyan},
24782 @item set style filename intensity @var{value}
24783 Set the intensity to @var{value}. Valid intensities are @samp{normal}
24784 (the default), @samp{bold}, and @samp{dim}.
24787 The style-able objects are:
24790 Control the styling of file names. By default, this style's
24791 foreground color is green.
24794 Control the styling of function names. These are managed with the
24795 @code{set style function} family of commands. By default, this
24796 style's foreground color is yellow.
24799 Control the styling of variable names. These are managed with the
24800 @code{set style variable} family of commands. By default, this style's
24801 foreground color is cyan.
24804 Control the styling of addresses. These are managed with the
24805 @code{set style address} family of commands. By default, this style's
24806 foreground color is blue.
24811 @cindex number representation
24812 @cindex entering numbers
24814 You can always enter numbers in octal, decimal, or hexadecimal in
24815 @value{GDBN} by the usual conventions: octal numbers begin with
24816 @samp{0}, decimal numbers end with @samp{.}, and hexadecimal numbers
24817 begin with @samp{0x}. Numbers that neither begin with @samp{0} or
24818 @samp{0x}, nor end with a @samp{.} are, by default, entered in base
24819 10; likewise, the default display for numbers---when no particular
24820 format is specified---is base 10. You can change the default base for
24821 both input and output with the commands described below.
24824 @kindex set input-radix
24825 @item set input-radix @var{base}
24826 Set the default base for numeric input. Supported choices
24827 for @var{base} are decimal 8, 10, or 16. The base must itself be
24828 specified either unambiguously or using the current input radix; for
24832 set input-radix 012
24833 set input-radix 10.
24834 set input-radix 0xa
24838 sets the input base to decimal. On the other hand, @samp{set input-radix 10}
24839 leaves the input radix unchanged, no matter what it was, since
24840 @samp{10}, being without any leading or trailing signs of its base, is
24841 interpreted in the current radix. Thus, if the current radix is 16,
24842 @samp{10} is interpreted in hex, i.e.@: as 16 decimal, which doesn't
24845 @kindex set output-radix
24846 @item set output-radix @var{base}
24847 Set the default base for numeric display. Supported choices
24848 for @var{base} are decimal 8, 10, or 16. The base must itself be
24849 specified either unambiguously or using the current input radix.
24851 @kindex show input-radix
24852 @item show input-radix
24853 Display the current default base for numeric input.
24855 @kindex show output-radix
24856 @item show output-radix
24857 Display the current default base for numeric display.
24859 @item set radix @r{[}@var{base}@r{]}
24863 These commands set and show the default base for both input and output
24864 of numbers. @code{set radix} sets the radix of input and output to
24865 the same base; without an argument, it resets the radix back to its
24866 default value of 10.
24871 @section Configuring the Current ABI
24873 @value{GDBN} can determine the @dfn{ABI} (Application Binary Interface) of your
24874 application automatically. However, sometimes you need to override its
24875 conclusions. Use these commands to manage @value{GDBN}'s view of the
24881 @cindex Newlib OS ABI and its influence on the longjmp handling
24883 One @value{GDBN} configuration can debug binaries for multiple operating
24884 system targets, either via remote debugging or native emulation.
24885 @value{GDBN} will autodetect the @dfn{OS ABI} (Operating System ABI) in use,
24886 but you can override its conclusion using the @code{set osabi} command.
24887 One example where this is useful is in debugging of binaries which use
24888 an alternate C library (e.g.@: @sc{uClibc} for @sc{gnu}/Linux) which does
24889 not have the same identifying marks that the standard C library for your
24892 When @value{GDBN} is debugging the AArch64 architecture, it provides a
24893 ``Newlib'' OS ABI. This is useful for handling @code{setjmp} and
24894 @code{longjmp} when debugging binaries that use the @sc{newlib} C library.
24895 The ``Newlib'' OS ABI can be selected by @code{set osabi Newlib}.
24899 Show the OS ABI currently in use.
24902 With no argument, show the list of registered available OS ABI's.
24904 @item set osabi @var{abi}
24905 Set the current OS ABI to @var{abi}.
24908 @cindex float promotion
24910 Generally, the way that an argument of type @code{float} is passed to a
24911 function depends on whether the function is prototyped. For a prototyped
24912 (i.e.@: ANSI/ISO style) function, @code{float} arguments are passed unchanged,
24913 according to the architecture's convention for @code{float}. For unprototyped
24914 (i.e.@: K&R style) functions, @code{float} arguments are first promoted to type
24915 @code{double} and then passed.
24917 Unfortunately, some forms of debug information do not reliably indicate whether
24918 a function is prototyped. If @value{GDBN} calls a function that is not marked
24919 as prototyped, it consults @kbd{set coerce-float-to-double}.
24922 @kindex set coerce-float-to-double
24923 @item set coerce-float-to-double
24924 @itemx set coerce-float-to-double on
24925 Arguments of type @code{float} will be promoted to @code{double} when passed
24926 to an unprototyped function. This is the default setting.
24928 @item set coerce-float-to-double off
24929 Arguments of type @code{float} will be passed directly to unprototyped
24932 @kindex show coerce-float-to-double
24933 @item show coerce-float-to-double
24934 Show the current setting of promoting @code{float} to @code{double}.
24938 @kindex show cp-abi
24939 @value{GDBN} needs to know the ABI used for your program's C@t{++}
24940 objects. The correct C@t{++} ABI depends on which C@t{++} compiler was
24941 used to build your application. @value{GDBN} only fully supports
24942 programs with a single C@t{++} ABI; if your program contains code using
24943 multiple C@t{++} ABI's or if @value{GDBN} can not identify your
24944 program's ABI correctly, you can tell @value{GDBN} which ABI to use.
24945 Currently supported ABI's include ``gnu-v2'', for @code{g++} versions
24946 before 3.0, ``gnu-v3'', for @code{g++} versions 3.0 and later, and
24947 ``hpaCC'' for the HP ANSI C@t{++} compiler. Other C@t{++} compilers may
24948 use the ``gnu-v2'' or ``gnu-v3'' ABI's as well. The default setting is
24953 Show the C@t{++} ABI currently in use.
24956 With no argument, show the list of supported C@t{++} ABI's.
24958 @item set cp-abi @var{abi}
24959 @itemx set cp-abi auto
24960 Set the current C@t{++} ABI to @var{abi}, or return to automatic detection.
24964 @section Automatically loading associated files
24965 @cindex auto-loading
24967 @value{GDBN} sometimes reads files with commands and settings automatically,
24968 without being explicitly told so by the user. We call this feature
24969 @dfn{auto-loading}. While auto-loading is useful for automatically adapting
24970 @value{GDBN} to the needs of your project, it can sometimes produce unexpected
24971 results or introduce security risks (e.g., if the file comes from untrusted
24975 * Init File in the Current Directory:: @samp{set/show/info auto-load local-gdbinit}
24976 * libthread_db.so.1 file:: @samp{set/show/info auto-load libthread-db}
24978 * Auto-loading safe path:: @samp{set/show/info auto-load safe-path}
24979 * Auto-loading verbose mode:: @samp{set/show debug auto-load}
24982 There are various kinds of files @value{GDBN} can automatically load.
24983 In addition to these files, @value{GDBN} supports auto-loading code written
24984 in various extension languages. @xref{Auto-loading extensions}.
24986 Note that loading of these associated files (including the local @file{.gdbinit}
24987 file) requires accordingly configured @code{auto-load safe-path}
24988 (@pxref{Auto-loading safe path}).
24990 For these reasons, @value{GDBN} includes commands and options to let you
24991 control when to auto-load files and which files should be auto-loaded.
24994 @anchor{set auto-load off}
24995 @kindex set auto-load off
24996 @item set auto-load off
24997 Globally disable loading of all auto-loaded files.
24998 You may want to use this command with the @samp{-iex} option
24999 (@pxref{Option -init-eval-command}) such as:
25001 $ @kbd{gdb -iex "set auto-load off" untrusted-executable corefile}
25004 Be aware that system init file (@pxref{System-wide configuration})
25005 and init files from your home directory (@pxref{Home Directory Init File})
25006 still get read (as they come from generally trusted directories).
25007 To prevent @value{GDBN} from auto-loading even those init files, use the
25008 @option{-nx} option (@pxref{Mode Options}), in addition to
25009 @code{set auto-load no}.
25011 @anchor{show auto-load}
25012 @kindex show auto-load
25013 @item show auto-load
25014 Show whether auto-loading of each specific @samp{auto-load} file(s) is enabled
25018 (gdb) show auto-load
25019 gdb-scripts: Auto-loading of canned sequences of commands scripts is on.
25020 libthread-db: Auto-loading of inferior specific libthread_db is on.
25021 local-gdbinit: Auto-loading of .gdbinit script from current directory
25023 python-scripts: Auto-loading of Python scripts is on.
25024 safe-path: List of directories from which it is safe to auto-load files
25025 is $debugdir:$datadir/auto-load.
25026 scripts-directory: List of directories from which to load auto-loaded scripts
25027 is $debugdir:$datadir/auto-load.
25030 @anchor{info auto-load}
25031 @kindex info auto-load
25032 @item info auto-load
25033 Print whether each specific @samp{auto-load} file(s) have been auto-loaded or
25037 (gdb) info auto-load
25040 Yes /home/user/gdb/gdb-gdb.gdb
25041 libthread-db: No auto-loaded libthread-db.
25042 local-gdbinit: Local .gdbinit file "/home/user/gdb/.gdbinit" has been
25046 Yes /home/user/gdb/gdb-gdb.py
25050 These are @value{GDBN} control commands for the auto-loading:
25052 @multitable @columnfractions .5 .5
25053 @item @xref{set auto-load off}.
25054 @tab Disable auto-loading globally.
25055 @item @xref{show auto-load}.
25056 @tab Show setting of all kinds of files.
25057 @item @xref{info auto-load}.
25058 @tab Show state of all kinds of files.
25059 @item @xref{set auto-load gdb-scripts}.
25060 @tab Control for @value{GDBN} command scripts.
25061 @item @xref{show auto-load gdb-scripts}.
25062 @tab Show setting of @value{GDBN} command scripts.
25063 @item @xref{info auto-load gdb-scripts}.
25064 @tab Show state of @value{GDBN} command scripts.
25065 @item @xref{set auto-load python-scripts}.
25066 @tab Control for @value{GDBN} Python scripts.
25067 @item @xref{show auto-load python-scripts}.
25068 @tab Show setting of @value{GDBN} Python scripts.
25069 @item @xref{info auto-load python-scripts}.
25070 @tab Show state of @value{GDBN} Python scripts.
25071 @item @xref{set auto-load guile-scripts}.
25072 @tab Control for @value{GDBN} Guile scripts.
25073 @item @xref{show auto-load guile-scripts}.
25074 @tab Show setting of @value{GDBN} Guile scripts.
25075 @item @xref{info auto-load guile-scripts}.
25076 @tab Show state of @value{GDBN} Guile scripts.
25077 @item @xref{set auto-load scripts-directory}.
25078 @tab Control for @value{GDBN} auto-loaded scripts location.
25079 @item @xref{show auto-load scripts-directory}.
25080 @tab Show @value{GDBN} auto-loaded scripts location.
25081 @item @xref{add-auto-load-scripts-directory}.
25082 @tab Add directory for auto-loaded scripts location list.
25083 @item @xref{set auto-load local-gdbinit}.
25084 @tab Control for init file in the current directory.
25085 @item @xref{show auto-load local-gdbinit}.
25086 @tab Show setting of init file in the current directory.
25087 @item @xref{info auto-load local-gdbinit}.
25088 @tab Show state of init file in the current directory.
25089 @item @xref{set auto-load libthread-db}.
25090 @tab Control for thread debugging library.
25091 @item @xref{show auto-load libthread-db}.
25092 @tab Show setting of thread debugging library.
25093 @item @xref{info auto-load libthread-db}.
25094 @tab Show state of thread debugging library.
25095 @item @xref{set auto-load safe-path}.
25096 @tab Control directories trusted for automatic loading.
25097 @item @xref{show auto-load safe-path}.
25098 @tab Show directories trusted for automatic loading.
25099 @item @xref{add-auto-load-safe-path}.
25100 @tab Add directory trusted for automatic loading.
25103 @node Init File in the Current Directory
25104 @subsection Automatically loading init file in the current directory
25105 @cindex auto-loading init file in the current directory
25107 By default, @value{GDBN} reads and executes the canned sequences of commands
25108 from init file (if any) in the current working directory,
25109 see @ref{Init File in the Current Directory during Startup}.
25111 Note that loading of this local @file{.gdbinit} file also requires accordingly
25112 configured @code{auto-load safe-path} (@pxref{Auto-loading safe path}).
25115 @anchor{set auto-load local-gdbinit}
25116 @kindex set auto-load local-gdbinit
25117 @item set auto-load local-gdbinit [on|off]
25118 Enable or disable the auto-loading of canned sequences of commands
25119 (@pxref{Sequences}) found in init file in the current directory.
25121 @anchor{show auto-load local-gdbinit}
25122 @kindex show auto-load local-gdbinit
25123 @item show auto-load local-gdbinit
25124 Show whether auto-loading of canned sequences of commands from init file in the
25125 current directory is enabled or disabled.
25127 @anchor{info auto-load local-gdbinit}
25128 @kindex info auto-load local-gdbinit
25129 @item info auto-load local-gdbinit
25130 Print whether canned sequences of commands from init file in the
25131 current directory have been auto-loaded.
25134 @node libthread_db.so.1 file
25135 @subsection Automatically loading thread debugging library
25136 @cindex auto-loading libthread_db.so.1
25138 This feature is currently present only on @sc{gnu}/Linux native hosts.
25140 @value{GDBN} reads in some cases thread debugging library from places specific
25141 to the inferior (@pxref{set libthread-db-search-path}).
25143 The special @samp{libthread-db-search-path} entry @samp{$sdir} is processed
25144 without checking this @samp{set auto-load libthread-db} switch as system
25145 libraries have to be trusted in general. In all other cases of
25146 @samp{libthread-db-search-path} entries @value{GDBN} checks first if @samp{set
25147 auto-load libthread-db} is enabled before trying to open such thread debugging
25150 Note that loading of this debugging library also requires accordingly configured
25151 @code{auto-load safe-path} (@pxref{Auto-loading safe path}).
25154 @anchor{set auto-load libthread-db}
25155 @kindex set auto-load libthread-db
25156 @item set auto-load libthread-db [on|off]
25157 Enable or disable the auto-loading of inferior specific thread debugging library.
25159 @anchor{show auto-load libthread-db}
25160 @kindex show auto-load libthread-db
25161 @item show auto-load libthread-db
25162 Show whether auto-loading of inferior specific thread debugging library is
25163 enabled or disabled.
25165 @anchor{info auto-load libthread-db}
25166 @kindex info auto-load libthread-db
25167 @item info auto-load libthread-db
25168 Print the list of all loaded inferior specific thread debugging libraries and
25169 for each such library print list of inferior @var{pid}s using it.
25172 @node Auto-loading safe path
25173 @subsection Security restriction for auto-loading
25174 @cindex auto-loading safe-path
25176 As the files of inferior can come from untrusted source (such as submitted by
25177 an application user) @value{GDBN} does not always load any files automatically.
25178 @value{GDBN} provides the @samp{set auto-load safe-path} setting to list
25179 directories trusted for loading files not explicitly requested by user.
25180 Each directory can also be a shell wildcard pattern.
25182 If the path is not set properly you will see a warning and the file will not
25187 Reading symbols from /home/user/gdb/gdb...done.
25188 warning: File "/home/user/gdb/gdb-gdb.gdb" auto-loading has been
25189 declined by your `auto-load safe-path' set
25190 to "$debugdir:$datadir/auto-load".
25191 warning: File "/home/user/gdb/gdb-gdb.py" auto-loading has been
25192 declined by your `auto-load safe-path' set
25193 to "$debugdir:$datadir/auto-load".
25197 To instruct @value{GDBN} to go ahead and use the init files anyway,
25198 invoke @value{GDBN} like this:
25201 $ gdb -q -iex "set auto-load safe-path /home/user/gdb" ./gdb
25204 The list of trusted directories is controlled by the following commands:
25207 @anchor{set auto-load safe-path}
25208 @kindex set auto-load safe-path
25209 @item set auto-load safe-path @r{[}@var{directories}@r{]}
25210 Set the list of directories (and their subdirectories) trusted for automatic
25211 loading and execution of scripts. You can also enter a specific trusted file.
25212 Each directory can also be a shell wildcard pattern; wildcards do not match
25213 directory separator - see @code{FNM_PATHNAME} for system function @code{fnmatch}
25214 (@pxref{Wildcard Matching, fnmatch, , libc, GNU C Library Reference Manual}).
25215 If you omit @var{directories}, @samp{auto-load safe-path} will be reset to
25216 its default value as specified during @value{GDBN} compilation.
25218 The list of directories uses path separator (@samp{:} on GNU and Unix
25219 systems, @samp{;} on MS-Windows and MS-DOS) to separate directories, similarly
25220 to the @env{PATH} environment variable.
25222 @anchor{show auto-load safe-path}
25223 @kindex show auto-load safe-path
25224 @item show auto-load safe-path
25225 Show the list of directories trusted for automatic loading and execution of
25228 @anchor{add-auto-load-safe-path}
25229 @kindex add-auto-load-safe-path
25230 @item add-auto-load-safe-path
25231 Add an entry (or list of entries) to the list of directories trusted for
25232 automatic loading and execution of scripts. Multiple entries may be delimited
25233 by the host platform path separator in use.
25236 This variable defaults to what @code{--with-auto-load-dir} has been configured
25237 to (@pxref{with-auto-load-dir}). @file{$debugdir} and @file{$datadir}
25238 substitution applies the same as for @ref{set auto-load scripts-directory}.
25239 The default @code{set auto-load safe-path} value can be also overriden by
25240 @value{GDBN} configuration option @option{--with-auto-load-safe-path}.
25242 Setting this variable to @file{/} disables this security protection,
25243 corresponding @value{GDBN} configuration option is
25244 @option{--without-auto-load-safe-path}.
25245 This variable is supposed to be set to the system directories writable by the
25246 system superuser only. Users can add their source directories in init files in
25247 their home directories (@pxref{Home Directory Init File}). See also deprecated
25248 init file in the current directory
25249 (@pxref{Init File in the Current Directory during Startup}).
25251 To force @value{GDBN} to load the files it declined to load in the previous
25252 example, you could use one of the following ways:
25255 @item @file{~/.gdbinit}: @samp{add-auto-load-safe-path ~/src/gdb}
25256 Specify this trusted directory (or a file) as additional component of the list.
25257 You have to specify also any existing directories displayed by
25258 by @samp{show auto-load safe-path} (such as @samp{/usr:/bin} in this example).
25260 @item @kbd{gdb -iex "set auto-load safe-path /usr:/bin:~/src/gdb" @dots{}}
25261 Specify this directory as in the previous case but just for a single
25262 @value{GDBN} session.
25264 @item @kbd{gdb -iex "set auto-load safe-path /" @dots{}}
25265 Disable auto-loading safety for a single @value{GDBN} session.
25266 This assumes all the files you debug during this @value{GDBN} session will come
25267 from trusted sources.
25269 @item @kbd{./configure --without-auto-load-safe-path}
25270 During compilation of @value{GDBN} you may disable any auto-loading safety.
25271 This assumes all the files you will ever debug with this @value{GDBN} come from
25275 On the other hand you can also explicitly forbid automatic files loading which
25276 also suppresses any such warning messages:
25279 @item @kbd{gdb -iex "set auto-load no" @dots{}}
25280 You can use @value{GDBN} command-line option for a single @value{GDBN} session.
25282 @item @file{~/.gdbinit}: @samp{set auto-load no}
25283 Disable auto-loading globally for the user
25284 (@pxref{Home Directory Init File}). While it is improbable, you could also
25285 use system init file instead (@pxref{System-wide configuration}).
25288 This setting applies to the file names as entered by user. If no entry matches
25289 @value{GDBN} tries as a last resort to also resolve all the file names into
25290 their canonical form (typically resolving symbolic links) and compare the
25291 entries again. @value{GDBN} already canonicalizes most of the filenames on its
25292 own before starting the comparison so a canonical form of directories is
25293 recommended to be entered.
25295 @node Auto-loading verbose mode
25296 @subsection Displaying files tried for auto-load
25297 @cindex auto-loading verbose mode
25299 For better visibility of all the file locations where you can place scripts to
25300 be auto-loaded with inferior --- or to protect yourself against accidental
25301 execution of untrusted scripts --- @value{GDBN} provides a feature for printing
25302 all the files attempted to be loaded. Both existing and non-existing files may
25305 For example the list of directories from which it is safe to auto-load files
25306 (@pxref{Auto-loading safe path}) applies also to canonicalized filenames which
25307 may not be too obvious while setting it up.
25310 (gdb) set debug auto-load on
25311 (gdb) file ~/src/t/true
25312 auto-load: Loading canned sequences of commands script "/tmp/true-gdb.gdb"
25313 for objfile "/tmp/true".
25314 auto-load: Updating directories of "/usr:/opt".
25315 auto-load: Using directory "/usr".
25316 auto-load: Using directory "/opt".
25317 warning: File "/tmp/true-gdb.gdb" auto-loading has been declined
25318 by your `auto-load safe-path' set to "/usr:/opt".
25322 @anchor{set debug auto-load}
25323 @kindex set debug auto-load
25324 @item set debug auto-load [on|off]
25325 Set whether to print the filenames attempted to be auto-loaded.
25327 @anchor{show debug auto-load}
25328 @kindex show debug auto-load
25329 @item show debug auto-load
25330 Show whether printing of the filenames attempted to be auto-loaded is turned
25334 @node Messages/Warnings
25335 @section Optional Warnings and Messages
25337 @cindex verbose operation
25338 @cindex optional warnings
25339 By default, @value{GDBN} is silent about its inner workings. If you are
25340 running on a slow machine, you may want to use the @code{set verbose}
25341 command. This makes @value{GDBN} tell you when it does a lengthy
25342 internal operation, so you will not think it has crashed.
25344 Currently, the messages controlled by @code{set verbose} are those
25345 which announce that the symbol table for a source file is being read;
25346 see @code{symbol-file} in @ref{Files, ,Commands to Specify Files}.
25349 @kindex set verbose
25350 @item set verbose on
25351 Enables @value{GDBN} output of certain informational messages.
25353 @item set verbose off
25354 Disables @value{GDBN} output of certain informational messages.
25356 @kindex show verbose
25358 Displays whether @code{set verbose} is on or off.
25361 By default, if @value{GDBN} encounters bugs in the symbol table of an
25362 object file, it is silent; but if you are debugging a compiler, you may
25363 find this information useful (@pxref{Symbol Errors, ,Errors Reading
25368 @kindex set complaints
25369 @item set complaints @var{limit}
25370 Permits @value{GDBN} to output @var{limit} complaints about each type of
25371 unusual symbols before becoming silent about the problem. Set
25372 @var{limit} to zero to suppress all complaints; set it to a large number
25373 to prevent complaints from being suppressed.
25375 @kindex show complaints
25376 @item show complaints
25377 Displays how many symbol complaints @value{GDBN} is permitted to produce.
25381 @anchor{confirmation requests}
25382 By default, @value{GDBN} is cautious, and asks what sometimes seems to be a
25383 lot of stupid questions to confirm certain commands. For example, if
25384 you try to run a program which is already running:
25388 The program being debugged has been started already.
25389 Start it from the beginning? (y or n)
25392 If you are willing to unflinchingly face the consequences of your own
25393 commands, you can disable this ``feature'':
25397 @kindex set confirm
25399 @cindex confirmation
25400 @cindex stupid questions
25401 @item set confirm off
25402 Disables confirmation requests. Note that running @value{GDBN} with
25403 the @option{--batch} option (@pxref{Mode Options, -batch}) also
25404 automatically disables confirmation requests.
25406 @item set confirm on
25407 Enables confirmation requests (the default).
25409 @kindex show confirm
25411 Displays state of confirmation requests.
25415 @cindex command tracing
25416 If you need to debug user-defined commands or sourced files you may find it
25417 useful to enable @dfn{command tracing}. In this mode each command will be
25418 printed as it is executed, prefixed with one or more @samp{+} symbols, the
25419 quantity denoting the call depth of each command.
25422 @kindex set trace-commands
25423 @cindex command scripts, debugging
25424 @item set trace-commands on
25425 Enable command tracing.
25426 @item set trace-commands off
25427 Disable command tracing.
25428 @item show trace-commands
25429 Display the current state of command tracing.
25432 @node Debugging Output
25433 @section Optional Messages about Internal Happenings
25434 @cindex optional debugging messages
25436 @value{GDBN} has commands that enable optional debugging messages from
25437 various @value{GDBN} subsystems; normally these commands are of
25438 interest to @value{GDBN} maintainers, or when reporting a bug. This
25439 section documents those commands.
25442 @kindex set exec-done-display
25443 @item set exec-done-display
25444 Turns on or off the notification of asynchronous commands'
25445 completion. When on, @value{GDBN} will print a message when an
25446 asynchronous command finishes its execution. The default is off.
25447 @kindex show exec-done-display
25448 @item show exec-done-display
25449 Displays the current setting of asynchronous command completion
25452 @cindex ARM AArch64
25453 @item set debug aarch64
25454 Turns on or off display of debugging messages related to ARM AArch64.
25455 The default is off.
25457 @item show debug aarch64
25458 Displays the current state of displaying debugging messages related to
25460 @cindex gdbarch debugging info
25461 @cindex architecture debugging info
25462 @item set debug arch
25463 Turns on or off display of gdbarch debugging info. The default is off
25464 @item show debug arch
25465 Displays the current state of displaying gdbarch debugging info.
25466 @item set debug aix-solib
25467 @cindex AIX shared library debugging
25468 Control display of debugging messages from the AIX shared library
25469 support module. The default is off.
25470 @item show debug aix-thread
25471 Show the current state of displaying AIX shared library debugging messages.
25472 @item set debug aix-thread
25473 @cindex AIX threads
25474 Display debugging messages about inner workings of the AIX thread
25476 @item show debug aix-thread
25477 Show the current state of AIX thread debugging info display.
25478 @item set debug check-physname
25480 Check the results of the ``physname'' computation. When reading DWARF
25481 debugging information for C@t{++}, @value{GDBN} attempts to compute
25482 each entity's name. @value{GDBN} can do this computation in two
25483 different ways, depending on exactly what information is present.
25484 When enabled, this setting causes @value{GDBN} to compute the names
25485 both ways and display any discrepancies.
25486 @item show debug check-physname
25487 Show the current state of ``physname'' checking.
25488 @item set debug coff-pe-read
25489 @cindex COFF/PE exported symbols
25490 Control display of debugging messages related to reading of COFF/PE
25491 exported symbols. The default is off.
25492 @item show debug coff-pe-read
25493 Displays the current state of displaying debugging messages related to
25494 reading of COFF/PE exported symbols.
25495 @item set debug dwarf-die
25497 Dump DWARF DIEs after they are read in.
25498 The value is the number of nesting levels to print.
25499 A value of zero turns off the display.
25500 @item show debug dwarf-die
25501 Show the current state of DWARF DIE debugging.
25502 @item set debug dwarf-line
25503 @cindex DWARF Line Tables
25504 Turns on or off display of debugging messages related to reading
25505 DWARF line tables. The default is 0 (off).
25506 A value of 1 provides basic information.
25507 A value greater than 1 provides more verbose information.
25508 @item show debug dwarf-line
25509 Show the current state of DWARF line table debugging.
25510 @item set debug dwarf-read
25511 @cindex DWARF Reading
25512 Turns on or off display of debugging messages related to reading
25513 DWARF debug info. The default is 0 (off).
25514 A value of 1 provides basic information.
25515 A value greater than 1 provides more verbose information.
25516 @item show debug dwarf-read
25517 Show the current state of DWARF reader debugging.
25518 @item set debug displaced
25519 @cindex displaced stepping debugging info
25520 Turns on or off display of @value{GDBN} debugging info for the
25521 displaced stepping support. The default is off.
25522 @item show debug displaced
25523 Displays the current state of displaying @value{GDBN} debugging info
25524 related to displaced stepping.
25525 @item set debug event
25526 @cindex event debugging info
25527 Turns on or off display of @value{GDBN} event debugging info. The
25529 @item show debug event
25530 Displays the current state of displaying @value{GDBN} event debugging
25532 @item set debug expression
25533 @cindex expression debugging info
25534 Turns on or off display of debugging info about @value{GDBN}
25535 expression parsing. The default is off.
25536 @item show debug expression
25537 Displays the current state of displaying debugging info about
25538 @value{GDBN} expression parsing.
25539 @item set debug fbsd-lwp
25540 @cindex FreeBSD LWP debug messages
25541 Turns on or off debugging messages from the FreeBSD LWP debug support.
25542 @item show debug fbsd-lwp
25543 Show the current state of FreeBSD LWP debugging messages.
25544 @item set debug fbsd-nat
25545 @cindex FreeBSD native target debug messages
25546 Turns on or off debugging messages from the FreeBSD native target.
25547 @item show debug fbsd-nat
25548 Show the current state of FreeBSD native target debugging messages.
25549 @item set debug frame
25550 @cindex frame debugging info
25551 Turns on or off display of @value{GDBN} frame debugging info. The
25553 @item show debug frame
25554 Displays the current state of displaying @value{GDBN} frame debugging
25556 @item set debug gnu-nat
25557 @cindex @sc{gnu}/Hurd debug messages
25558 Turn on or off debugging messages from the @sc{gnu}/Hurd debug support.
25559 @item show debug gnu-nat
25560 Show the current state of @sc{gnu}/Hurd debugging messages.
25561 @item set debug infrun
25562 @cindex inferior debugging info
25563 Turns on or off display of @value{GDBN} debugging info for running the inferior.
25564 The default is off. @file{infrun.c} contains GDB's runtime state machine used
25565 for implementing operations such as single-stepping the inferior.
25566 @item show debug infrun
25567 Displays the current state of @value{GDBN} inferior debugging.
25568 @item set debug jit
25569 @cindex just-in-time compilation, debugging messages
25570 Turn on or off debugging messages from JIT debug support.
25571 @item show debug jit
25572 Displays the current state of @value{GDBN} JIT debugging.
25573 @item set debug lin-lwp
25574 @cindex @sc{gnu}/Linux LWP debug messages
25575 @cindex Linux lightweight processes
25576 Turn on or off debugging messages from the Linux LWP debug support.
25577 @item show debug lin-lwp
25578 Show the current state of Linux LWP debugging messages.
25579 @item set debug linux-namespaces
25580 @cindex @sc{gnu}/Linux namespaces debug messages
25581 Turn on or off debugging messages from the Linux namespaces debug support.
25582 @item show debug linux-namespaces
25583 Show the current state of Linux namespaces debugging messages.
25584 @item set debug mach-o
25585 @cindex Mach-O symbols processing
25586 Control display of debugging messages related to Mach-O symbols
25587 processing. The default is off.
25588 @item show debug mach-o
25589 Displays the current state of displaying debugging messages related to
25590 reading of COFF/PE exported symbols.
25591 @item set debug notification
25592 @cindex remote async notification debugging info
25593 Turn on or off debugging messages about remote async notification.
25594 The default is off.
25595 @item show debug notification
25596 Displays the current state of remote async notification debugging messages.
25597 @item set debug observer
25598 @cindex observer debugging info
25599 Turns on or off display of @value{GDBN} observer debugging. This
25600 includes info such as the notification of observable events.
25601 @item show debug observer
25602 Displays the current state of observer debugging.
25603 @item set debug overload
25604 @cindex C@t{++} overload debugging info
25605 Turns on or off display of @value{GDBN} C@t{++} overload debugging
25606 info. This includes info such as ranking of functions, etc. The default
25608 @item show debug overload
25609 Displays the current state of displaying @value{GDBN} C@t{++} overload
25611 @cindex expression parser, debugging info
25612 @cindex debug expression parser
25613 @item set debug parser
25614 Turns on or off the display of expression parser debugging output.
25615 Internally, this sets the @code{yydebug} variable in the expression
25616 parser. @xref{Tracing, , Tracing Your Parser, bison, Bison}, for
25617 details. The default is off.
25618 @item show debug parser
25619 Show the current state of expression parser debugging.
25620 @cindex packets, reporting on stdout
25621 @cindex serial connections, debugging
25622 @cindex debug remote protocol
25623 @cindex remote protocol debugging
25624 @cindex display remote packets
25625 @item set debug remote
25626 Turns on or off display of reports on all packets sent back and forth across
25627 the serial line to the remote machine. The info is printed on the
25628 @value{GDBN} standard output stream. The default is off.
25629 @item show debug remote
25630 Displays the state of display of remote packets.
25632 @item set debug separate-debug-file
25633 Turns on or off display of debug output about separate debug file search.
25634 @item show debug separate-debug-file
25635 Displays the state of separate debug file search debug output.
25637 @item set debug serial
25638 Turns on or off display of @value{GDBN} serial debugging info. The
25640 @item show debug serial
25641 Displays the current state of displaying @value{GDBN} serial debugging
25643 @item set debug solib-frv
25644 @cindex FR-V shared-library debugging
25645 Turn on or off debugging messages for FR-V shared-library code.
25646 @item show debug solib-frv
25647 Display the current state of FR-V shared-library code debugging
25649 @item set debug symbol-lookup
25650 @cindex symbol lookup
25651 Turns on or off display of debugging messages related to symbol lookup.
25652 The default is 0 (off).
25653 A value of 1 provides basic information.
25654 A value greater than 1 provides more verbose information.
25655 @item show debug symbol-lookup
25656 Show the current state of symbol lookup debugging messages.
25657 @item set debug symfile
25658 @cindex symbol file functions
25659 Turns on or off display of debugging messages related to symbol file functions.
25660 The default is off. @xref{Files}.
25661 @item show debug symfile
25662 Show the current state of symbol file debugging messages.
25663 @item set debug symtab-create
25664 @cindex symbol table creation
25665 Turns on or off display of debugging messages related to symbol table creation.
25666 The default is 0 (off).
25667 A value of 1 provides basic information.
25668 A value greater than 1 provides more verbose information.
25669 @item show debug symtab-create
25670 Show the current state of symbol table creation debugging.
25671 @item set debug target
25672 @cindex target debugging info
25673 Turns on or off display of @value{GDBN} target debugging info. This info
25674 includes what is going on at the target level of GDB, as it happens. The
25675 default is 0. Set it to 1 to track events, and to 2 to also track the
25676 value of large memory transfers.
25677 @item show debug target
25678 Displays the current state of displaying @value{GDBN} target debugging
25680 @item set debug timestamp
25681 @cindex timestampping debugging info
25682 Turns on or off display of timestamps with @value{GDBN} debugging info.
25683 When enabled, seconds and microseconds are displayed before each debugging
25685 @item show debug timestamp
25686 Displays the current state of displaying timestamps with @value{GDBN}
25688 @item set debug varobj
25689 @cindex variable object debugging info
25690 Turns on or off display of @value{GDBN} variable object debugging
25691 info. The default is off.
25692 @item show debug varobj
25693 Displays the current state of displaying @value{GDBN} variable object
25695 @item set debug xml
25696 @cindex XML parser debugging
25697 Turn on or off debugging messages for built-in XML parsers.
25698 @item show debug xml
25699 Displays the current state of XML debugging messages.
25702 @node Other Misc Settings
25703 @section Other Miscellaneous Settings
25704 @cindex miscellaneous settings
25707 @kindex set interactive-mode
25708 @item set interactive-mode
25709 If @code{on}, forces @value{GDBN} to assume that GDB was started
25710 in a terminal. In practice, this means that @value{GDBN} should wait
25711 for the user to answer queries generated by commands entered at
25712 the command prompt. If @code{off}, forces @value{GDBN} to operate
25713 in the opposite mode, and it uses the default answers to all queries.
25714 If @code{auto} (the default), @value{GDBN} tries to determine whether
25715 its standard input is a terminal, and works in interactive-mode if it
25716 is, non-interactively otherwise.
25718 In the vast majority of cases, the debugger should be able to guess
25719 correctly which mode should be used. But this setting can be useful
25720 in certain specific cases, such as running a MinGW @value{GDBN}
25721 inside a cygwin window.
25723 @kindex show interactive-mode
25724 @item show interactive-mode
25725 Displays whether the debugger is operating in interactive mode or not.
25728 @node Extending GDB
25729 @chapter Extending @value{GDBN}
25730 @cindex extending GDB
25732 @value{GDBN} provides several mechanisms for extension.
25733 @value{GDBN} also provides the ability to automatically load
25734 extensions when it reads a file for debugging. This allows the
25735 user to automatically customize @value{GDBN} for the program
25739 * Sequences:: Canned Sequences of @value{GDBN} Commands
25740 * Python:: Extending @value{GDBN} using Python
25741 * Guile:: Extending @value{GDBN} using Guile
25742 * Auto-loading extensions:: Automatically loading extensions
25743 * Multiple Extension Languages:: Working with multiple extension languages
25744 * Aliases:: Creating new spellings of existing commands
25747 To facilitate the use of extension languages, @value{GDBN} is capable
25748 of evaluating the contents of a file. When doing so, @value{GDBN}
25749 can recognize which extension language is being used by looking at
25750 the filename extension. Files with an unrecognized filename extension
25751 are always treated as a @value{GDBN} Command Files.
25752 @xref{Command Files,, Command files}.
25754 You can control how @value{GDBN} evaluates these files with the following
25758 @kindex set script-extension
25759 @kindex show script-extension
25760 @item set script-extension off
25761 All scripts are always evaluated as @value{GDBN} Command Files.
25763 @item set script-extension soft
25764 The debugger determines the scripting language based on filename
25765 extension. If this scripting language is supported, @value{GDBN}
25766 evaluates the script using that language. Otherwise, it evaluates
25767 the file as a @value{GDBN} Command File.
25769 @item set script-extension strict
25770 The debugger determines the scripting language based on filename
25771 extension, and evaluates the script using that language. If the
25772 language is not supported, then the evaluation fails.
25774 @item show script-extension
25775 Display the current value of the @code{script-extension} option.
25780 @section Canned Sequences of Commands
25782 Aside from breakpoint commands (@pxref{Break Commands, ,Breakpoint
25783 Command Lists}), @value{GDBN} provides two ways to store sequences of
25784 commands for execution as a unit: user-defined commands and command
25788 * Define:: How to define your own commands
25789 * Hooks:: Hooks for user-defined commands
25790 * Command Files:: How to write scripts of commands to be stored in a file
25791 * Output:: Commands for controlled output
25792 * Auto-loading sequences:: Controlling auto-loaded command files
25796 @subsection User-defined Commands
25798 @cindex user-defined command
25799 @cindex arguments, to user-defined commands
25800 A @dfn{user-defined command} is a sequence of @value{GDBN} commands to
25801 which you assign a new name as a command. This is done with the
25802 @code{define} command. User commands may accept an unlimited number of arguments
25803 separated by whitespace. Arguments are accessed within the user command
25804 via @code{$arg0@dots{}$argN}. A trivial example:
25808 print $arg0 + $arg1 + $arg2
25813 To execute the command use:
25820 This defines the command @code{adder}, which prints the sum of
25821 its three arguments. Note the arguments are text substitutions, so they may
25822 reference variables, use complex expressions, or even perform inferior
25825 @cindex argument count in user-defined commands
25826 @cindex how many arguments (user-defined commands)
25827 In addition, @code{$argc} may be used to find out how many arguments have
25833 print $arg0 + $arg1
25836 print $arg0 + $arg1 + $arg2
25841 Combining with the @code{eval} command (@pxref{eval}) makes it easier
25842 to process a variable number of arguments:
25849 eval "set $sum = $sum + $arg%d", $i
25859 @item define @var{commandname}
25860 Define a command named @var{commandname}. If there is already a command
25861 by that name, you are asked to confirm that you want to redefine it.
25862 The argument @var{commandname} may be a bare command name consisting of letters,
25863 numbers, dashes, and underscores. It may also start with any predefined
25864 prefix command. For example, @samp{define target my-target} creates
25865 a user-defined @samp{target my-target} command.
25867 The definition of the command is made up of other @value{GDBN} command lines,
25868 which are given following the @code{define} command. The end of these
25869 commands is marked by a line containing @code{end}.
25872 @kindex end@r{ (user-defined commands)}
25873 @item document @var{commandname}
25874 Document the user-defined command @var{commandname}, so that it can be
25875 accessed by @code{help}. The command @var{commandname} must already be
25876 defined. This command reads lines of documentation just as @code{define}
25877 reads the lines of the command definition, ending with @code{end}.
25878 After the @code{document} command is finished, @code{help} on command
25879 @var{commandname} displays the documentation you have written.
25881 You may use the @code{document} command again to change the
25882 documentation of a command. Redefining the command with @code{define}
25883 does not change the documentation.
25885 @kindex dont-repeat
25886 @cindex don't repeat command
25888 Used inside a user-defined command, this tells @value{GDBN} that this
25889 command should not be repeated when the user hits @key{RET}
25890 (@pxref{Command Syntax, repeat last command}).
25892 @kindex help user-defined
25893 @item help user-defined
25894 List all user-defined commands and all python commands defined in class
25895 COMAND_USER. The first line of the documentation or docstring is
25900 @itemx show user @var{commandname}
25901 Display the @value{GDBN} commands used to define @var{commandname} (but
25902 not its documentation). If no @var{commandname} is given, display the
25903 definitions for all user-defined commands.
25904 This does not work for user-defined python commands.
25906 @cindex infinite recursion in user-defined commands
25907 @kindex show max-user-call-depth
25908 @kindex set max-user-call-depth
25909 @item show max-user-call-depth
25910 @itemx set max-user-call-depth
25911 The value of @code{max-user-call-depth} controls how many recursion
25912 levels are allowed in user-defined commands before @value{GDBN} suspects an
25913 infinite recursion and aborts the command.
25914 This does not apply to user-defined python commands.
25917 In addition to the above commands, user-defined commands frequently
25918 use control flow commands, described in @ref{Command Files}.
25920 When user-defined commands are executed, the
25921 commands of the definition are not printed. An error in any command
25922 stops execution of the user-defined command.
25924 If used interactively, commands that would ask for confirmation proceed
25925 without asking when used inside a user-defined command. Many @value{GDBN}
25926 commands that normally print messages to say what they are doing omit the
25927 messages when used in a user-defined command.
25930 @subsection User-defined Command Hooks
25931 @cindex command hooks
25932 @cindex hooks, for commands
25933 @cindex hooks, pre-command
25936 You may define @dfn{hooks}, which are a special kind of user-defined
25937 command. Whenever you run the command @samp{foo}, if the user-defined
25938 command @samp{hook-foo} exists, it is executed (with no arguments)
25939 before that command.
25941 @cindex hooks, post-command
25943 A hook may also be defined which is run after the command you executed.
25944 Whenever you run the command @samp{foo}, if the user-defined command
25945 @samp{hookpost-foo} exists, it is executed (with no arguments) after
25946 that command. Post-execution hooks may exist simultaneously with
25947 pre-execution hooks, for the same command.
25949 It is valid for a hook to call the command which it hooks. If this
25950 occurs, the hook is not re-executed, thereby avoiding infinite recursion.
25952 @c It would be nice if hookpost could be passed a parameter indicating
25953 @c if the command it hooks executed properly or not. FIXME!
25955 @kindex stop@r{, a pseudo-command}
25956 In addition, a pseudo-command, @samp{stop} exists. Defining
25957 (@samp{hook-stop}) makes the associated commands execute every time
25958 execution stops in your program: before breakpoint commands are run,
25959 displays are printed, or the stack frame is printed.
25961 For example, to ignore @code{SIGALRM} signals while
25962 single-stepping, but treat them normally during normal execution,
25967 handle SIGALRM nopass
25971 handle SIGALRM pass
25974 define hook-continue
25975 handle SIGALRM pass
25979 As a further example, to hook at the beginning and end of the @code{echo}
25980 command, and to add extra text to the beginning and end of the message,
25988 define hookpost-echo
25992 (@value{GDBP}) echo Hello World
25993 <<<---Hello World--->>>
25998 You can define a hook for any single-word command in @value{GDBN}, but
25999 not for command aliases; you should define a hook for the basic command
26000 name, e.g.@: @code{backtrace} rather than @code{bt}.
26001 @c FIXME! So how does Joe User discover whether a command is an alias
26003 You can hook a multi-word command by adding @code{hook-} or
26004 @code{hookpost-} to the last word of the command, e.g.@:
26005 @samp{define target hook-remote} to add a hook to @samp{target remote}.
26007 If an error occurs during the execution of your hook, execution of
26008 @value{GDBN} commands stops and @value{GDBN} issues a prompt
26009 (before the command that you actually typed had a chance to run).
26011 If you try to define a hook which does not match any known command, you
26012 get a warning from the @code{define} command.
26014 @node Command Files
26015 @subsection Command Files
26017 @cindex command files
26018 @cindex scripting commands
26019 A command file for @value{GDBN} is a text file made of lines that are
26020 @value{GDBN} commands. Comments (lines starting with @kbd{#}) may
26021 also be included. An empty line in a command file does nothing; it
26022 does not mean to repeat the last command, as it would from the
26025 You can request the execution of a command file with the @code{source}
26026 command. Note that the @code{source} command is also used to evaluate
26027 scripts that are not Command Files. The exact behavior can be configured
26028 using the @code{script-extension} setting.
26029 @xref{Extending GDB,, Extending GDB}.
26033 @cindex execute commands from a file
26034 @item source [-s] [-v] @var{filename}
26035 Execute the command file @var{filename}.
26038 The lines in a command file are generally executed sequentially,
26039 unless the order of execution is changed by one of the
26040 @emph{flow-control commands} described below. The commands are not
26041 printed as they are executed. An error in any command terminates
26042 execution of the command file and control is returned to the console.
26044 @value{GDBN} first searches for @var{filename} in the current directory.
26045 If the file is not found there, and @var{filename} does not specify a
26046 directory, then @value{GDBN} also looks for the file on the source search path
26047 (specified with the @samp{directory} command);
26048 except that @file{$cdir} is not searched because the compilation directory
26049 is not relevant to scripts.
26051 If @code{-s} is specified, then @value{GDBN} searches for @var{filename}
26052 on the search path even if @var{filename} specifies a directory.
26053 The search is done by appending @var{filename} to each element of the
26054 search path. So, for example, if @var{filename} is @file{mylib/myscript}
26055 and the search path contains @file{/home/user} then @value{GDBN} will
26056 look for the script @file{/home/user/mylib/myscript}.
26057 The search is also done if @var{filename} is an absolute path.
26058 For example, if @var{filename} is @file{/tmp/myscript} and
26059 the search path contains @file{/home/user} then @value{GDBN} will
26060 look for the script @file{/home/user/tmp/myscript}.
26061 For DOS-like systems, if @var{filename} contains a drive specification,
26062 it is stripped before concatenation. For example, if @var{filename} is
26063 @file{d:myscript} and the search path contains @file{c:/tmp} then @value{GDBN}
26064 will look for the script @file{c:/tmp/myscript}.
26066 If @code{-v}, for verbose mode, is given then @value{GDBN} displays
26067 each command as it is executed. The option must be given before
26068 @var{filename}, and is interpreted as part of the filename anywhere else.
26070 Commands that would ask for confirmation if used interactively proceed
26071 without asking when used in a command file. Many @value{GDBN} commands that
26072 normally print messages to say what they are doing omit the messages
26073 when called from command files.
26075 @value{GDBN} also accepts command input from standard input. In this
26076 mode, normal output goes to standard output and error output goes to
26077 standard error. Errors in a command file supplied on standard input do
26078 not terminate execution of the command file---execution continues with
26082 gdb < cmds > log 2>&1
26085 (The syntax above will vary depending on the shell used.) This example
26086 will execute commands from the file @file{cmds}. All output and errors
26087 would be directed to @file{log}.
26089 Since commands stored on command files tend to be more general than
26090 commands typed interactively, they frequently need to deal with
26091 complicated situations, such as different or unexpected values of
26092 variables and symbols, changes in how the program being debugged is
26093 built, etc. @value{GDBN} provides a set of flow-control commands to
26094 deal with these complexities. Using these commands, you can write
26095 complex scripts that loop over data structures, execute commands
26096 conditionally, etc.
26103 This command allows to include in your script conditionally executed
26104 commands. The @code{if} command takes a single argument, which is an
26105 expression to evaluate. It is followed by a series of commands that
26106 are executed only if the expression is true (its value is nonzero).
26107 There can then optionally be an @code{else} line, followed by a series
26108 of commands that are only executed if the expression was false. The
26109 end of the list is marked by a line containing @code{end}.
26113 This command allows to write loops. Its syntax is similar to
26114 @code{if}: the command takes a single argument, which is an expression
26115 to evaluate, and must be followed by the commands to execute, one per
26116 line, terminated by an @code{end}. These commands are called the
26117 @dfn{body} of the loop. The commands in the body of @code{while} are
26118 executed repeatedly as long as the expression evaluates to true.
26122 This command exits the @code{while} loop in whose body it is included.
26123 Execution of the script continues after that @code{while}s @code{end}
26126 @kindex loop_continue
26127 @item loop_continue
26128 This command skips the execution of the rest of the body of commands
26129 in the @code{while} loop in whose body it is included. Execution
26130 branches to the beginning of the @code{while} loop, where it evaluates
26131 the controlling expression.
26133 @kindex end@r{ (if/else/while commands)}
26135 Terminate the block of commands that are the body of @code{if},
26136 @code{else}, or @code{while} flow-control commands.
26141 @subsection Commands for Controlled Output
26143 During the execution of a command file or a user-defined command, normal
26144 @value{GDBN} output is suppressed; the only output that appears is what is
26145 explicitly printed by the commands in the definition. This section
26146 describes three commands useful for generating exactly the output you
26151 @item echo @var{text}
26152 @c I do not consider backslash-space a standard C escape sequence
26153 @c because it is not in ANSI.
26154 Print @var{text}. Nonprinting characters can be included in
26155 @var{text} using C escape sequences, such as @samp{\n} to print a
26156 newline. @strong{No newline is printed unless you specify one.}
26157 In addition to the standard C escape sequences, a backslash followed
26158 by a space stands for a space. This is useful for displaying a
26159 string with spaces at the beginning or the end, since leading and
26160 trailing spaces are otherwise trimmed from all arguments.
26161 To print @samp{@w{ }and foo =@w{ }}, use the command
26162 @samp{echo \@w{ }and foo = \@w{ }}.
26164 A backslash at the end of @var{text} can be used, as in C, to continue
26165 the command onto subsequent lines. For example,
26168 echo This is some text\n\
26169 which is continued\n\
26170 onto several lines.\n
26173 produces the same output as
26176 echo This is some text\n
26177 echo which is continued\n
26178 echo onto several lines.\n
26182 @item output @var{expression}
26183 Print the value of @var{expression} and nothing but that value: no
26184 newlines, no @samp{$@var{nn} = }. The value is not entered in the
26185 value history either. @xref{Expressions, ,Expressions}, for more information
26188 @item output/@var{fmt} @var{expression}
26189 Print the value of @var{expression} in format @var{fmt}. You can use
26190 the same formats as for @code{print}. @xref{Output Formats,,Output
26191 Formats}, for more information.
26194 @item printf @var{template}, @var{expressions}@dots{}
26195 Print the values of one or more @var{expressions} under the control of
26196 the string @var{template}. To print several values, make
26197 @var{expressions} be a comma-separated list of individual expressions,
26198 which may be either numbers or pointers. Their values are printed as
26199 specified by @var{template}, exactly as a C program would do by
26200 executing the code below:
26203 printf (@var{template}, @var{expressions}@dots{});
26206 As in @code{C} @code{printf}, ordinary characters in @var{template}
26207 are printed verbatim, while @dfn{conversion specification} introduced
26208 by the @samp{%} character cause subsequent @var{expressions} to be
26209 evaluated, their values converted and formatted according to type and
26210 style information encoded in the conversion specifications, and then
26213 For example, you can print two values in hex like this:
26216 printf "foo, bar-foo = 0x%x, 0x%x\n", foo, bar-foo
26219 @code{printf} supports all the standard @code{C} conversion
26220 specifications, including the flags and modifiers between the @samp{%}
26221 character and the conversion letter, with the following exceptions:
26225 The argument-ordering modifiers, such as @samp{2$}, are not supported.
26228 The modifier @samp{*} is not supported for specifying precision or
26232 The @samp{'} flag (for separation of digits into groups according to
26233 @code{LC_NUMERIC'}) is not supported.
26236 The type modifiers @samp{hh}, @samp{j}, @samp{t}, and @samp{z} are not
26240 The conversion letter @samp{n} (as in @samp{%n}) is not supported.
26243 The conversion letters @samp{a} and @samp{A} are not supported.
26247 Note that the @samp{ll} type modifier is supported only if the
26248 underlying @code{C} implementation used to build @value{GDBN} supports
26249 the @code{long long int} type, and the @samp{L} type modifier is
26250 supported only if @code{long double} type is available.
26252 As in @code{C}, @code{printf} supports simple backslash-escape
26253 sequences, such as @code{\n}, @samp{\t}, @samp{\\}, @samp{\"},
26254 @samp{\a}, and @samp{\f}, that consist of backslash followed by a
26255 single character. Octal and hexadecimal escape sequences are not
26258 Additionally, @code{printf} supports conversion specifications for DFP
26259 (@dfn{Decimal Floating Point}) types using the following length modifiers
26260 together with a floating point specifier.
26265 @samp{H} for printing @code{Decimal32} types.
26268 @samp{D} for printing @code{Decimal64} types.
26271 @samp{DD} for printing @code{Decimal128} types.
26274 If the underlying @code{C} implementation used to build @value{GDBN} has
26275 support for the three length modifiers for DFP types, other modifiers
26276 such as width and precision will also be available for @value{GDBN} to use.
26278 In case there is no such @code{C} support, no additional modifiers will be
26279 available and the value will be printed in the standard way.
26281 Here's an example of printing DFP types using the above conversion letters:
26283 printf "D32: %Hf - D64: %Df - D128: %DDf\n",1.2345df,1.2E10dd,1.2E1dl
26288 @item eval @var{template}, @var{expressions}@dots{}
26289 Convert the values of one or more @var{expressions} under the control of
26290 the string @var{template} to a command line, and call it.
26294 @node Auto-loading sequences
26295 @subsection Controlling auto-loading native @value{GDBN} scripts
26296 @cindex native script auto-loading
26298 When a new object file is read (for example, due to the @code{file}
26299 command, or because the inferior has loaded a shared library),
26300 @value{GDBN} will look for the command file @file{@var{objfile}-gdb.gdb}.
26301 @xref{Auto-loading extensions}.
26303 Auto-loading can be enabled or disabled,
26304 and the list of auto-loaded scripts can be printed.
26307 @anchor{set auto-load gdb-scripts}
26308 @kindex set auto-load gdb-scripts
26309 @item set auto-load gdb-scripts [on|off]
26310 Enable or disable the auto-loading of canned sequences of commands scripts.
26312 @anchor{show auto-load gdb-scripts}
26313 @kindex show auto-load gdb-scripts
26314 @item show auto-load gdb-scripts
26315 Show whether auto-loading of canned sequences of commands scripts is enabled or
26318 @anchor{info auto-load gdb-scripts}
26319 @kindex info auto-load gdb-scripts
26320 @cindex print list of auto-loaded canned sequences of commands scripts
26321 @item info auto-load gdb-scripts [@var{regexp}]
26322 Print the list of all canned sequences of commands scripts that @value{GDBN}
26326 If @var{regexp} is supplied only canned sequences of commands scripts with
26327 matching names are printed.
26329 @c Python docs live in a separate file.
26330 @include python.texi
26332 @c Guile docs live in a separate file.
26333 @include guile.texi
26335 @node Auto-loading extensions
26336 @section Auto-loading extensions
26337 @cindex auto-loading extensions
26339 @value{GDBN} provides two mechanisms for automatically loading extensions
26340 when a new object file is read (for example, due to the @code{file}
26341 command, or because the inferior has loaded a shared library):
26342 @file{@var{objfile}-gdb.@var{ext}} and the @code{.debug_gdb_scripts}
26343 section of modern file formats like ELF.
26346 * objfile-gdb.ext file: objfile-gdbdotext file. The @file{@var{objfile}-gdb.@var{ext}} file
26347 * .debug_gdb_scripts section: dotdebug_gdb_scripts section. The @code{.debug_gdb_scripts} section
26348 * Which flavor to choose?::
26351 The auto-loading feature is useful for supplying application-specific
26352 debugging commands and features.
26354 Auto-loading can be enabled or disabled,
26355 and the list of auto-loaded scripts can be printed.
26356 See the @samp{auto-loading} section of each extension language
26357 for more information.
26358 For @value{GDBN} command files see @ref{Auto-loading sequences}.
26359 For Python files see @ref{Python Auto-loading}.
26361 Note that loading of this script file also requires accordingly configured
26362 @code{auto-load safe-path} (@pxref{Auto-loading safe path}).
26364 @node objfile-gdbdotext file
26365 @subsection The @file{@var{objfile}-gdb.@var{ext}} file
26366 @cindex @file{@var{objfile}-gdb.gdb}
26367 @cindex @file{@var{objfile}-gdb.py}
26368 @cindex @file{@var{objfile}-gdb.scm}
26370 When a new object file is read, @value{GDBN} looks for a file named
26371 @file{@var{objfile}-gdb.@var{ext}} (we call it @var{script-name} below),
26372 where @var{objfile} is the object file's name and
26373 where @var{ext} is the file extension for the extension language:
26376 @item @file{@var{objfile}-gdb.gdb}
26377 GDB's own command language
26378 @item @file{@var{objfile}-gdb.py}
26380 @item @file{@var{objfile}-gdb.scm}
26384 @var{script-name} is formed by ensuring that the file name of @var{objfile}
26385 is absolute, following all symlinks, and resolving @code{.} and @code{..}
26386 components, and appending the @file{-gdb.@var{ext}} suffix.
26387 If this file exists and is readable, @value{GDBN} will evaluate it as a
26388 script in the specified extension language.
26390 If this file does not exist, then @value{GDBN} will look for
26391 @var{script-name} file in all of the directories as specified below.
26393 Note that loading of these files requires an accordingly configured
26394 @code{auto-load safe-path} (@pxref{Auto-loading safe path}).
26396 For object files using @file{.exe} suffix @value{GDBN} tries to load first the
26397 scripts normally according to its @file{.exe} filename. But if no scripts are
26398 found @value{GDBN} also tries script filenames matching the object file without
26399 its @file{.exe} suffix. This @file{.exe} stripping is case insensitive and it
26400 is attempted on any platform. This makes the script filenames compatible
26401 between Unix and MS-Windows hosts.
26404 @anchor{set auto-load scripts-directory}
26405 @kindex set auto-load scripts-directory
26406 @item set auto-load scripts-directory @r{[}@var{directories}@r{]}
26407 Control @value{GDBN} auto-loaded scripts location. Multiple directory entries
26408 may be delimited by the host platform path separator in use
26409 (@samp{:} on Unix, @samp{;} on MS-Windows and MS-DOS).
26411 Each entry here needs to be covered also by the security setting
26412 @code{set auto-load safe-path} (@pxref{set auto-load safe-path}).
26414 @anchor{with-auto-load-dir}
26415 This variable defaults to @file{$debugdir:$datadir/auto-load}. The default
26416 @code{set auto-load safe-path} value can be also overriden by @value{GDBN}
26417 configuration option @option{--with-auto-load-dir}.
26419 Any reference to @file{$debugdir} will get replaced by
26420 @var{debug-file-directory} value (@pxref{Separate Debug Files}) and any
26421 reference to @file{$datadir} will get replaced by @var{data-directory} which is
26422 determined at @value{GDBN} startup (@pxref{Data Files}). @file{$debugdir} and
26423 @file{$datadir} must be placed as a directory component --- either alone or
26424 delimited by @file{/} or @file{\} directory separators, depending on the host
26427 The list of directories uses path separator (@samp{:} on GNU and Unix
26428 systems, @samp{;} on MS-Windows and MS-DOS) to separate directories, similarly
26429 to the @env{PATH} environment variable.
26431 @anchor{show auto-load scripts-directory}
26432 @kindex show auto-load scripts-directory
26433 @item show auto-load scripts-directory
26434 Show @value{GDBN} auto-loaded scripts location.
26436 @anchor{add-auto-load-scripts-directory}
26437 @kindex add-auto-load-scripts-directory
26438 @item add-auto-load-scripts-directory @r{[}@var{directories}@dots{}@r{]}
26439 Add an entry (or list of entries) to the list of auto-loaded scripts locations.
26440 Multiple entries may be delimited by the host platform path separator in use.
26443 @value{GDBN} does not track which files it has already auto-loaded this way.
26444 @value{GDBN} will load the associated script every time the corresponding
26445 @var{objfile} is opened.
26446 So your @file{-gdb.@var{ext}} file should be careful to avoid errors if it
26447 is evaluated more than once.
26449 @node dotdebug_gdb_scripts section
26450 @subsection The @code{.debug_gdb_scripts} section
26451 @cindex @code{.debug_gdb_scripts} section
26453 For systems using file formats like ELF and COFF,
26454 when @value{GDBN} loads a new object file
26455 it will look for a special section named @code{.debug_gdb_scripts}.
26456 If this section exists, its contents is a list of null-terminated entries
26457 specifying scripts to load. Each entry begins with a non-null prefix byte that
26458 specifies the kind of entry, typically the extension language and whether the
26459 script is in a file or inlined in @code{.debug_gdb_scripts}.
26461 The following entries are supported:
26464 @item SECTION_SCRIPT_ID_PYTHON_FILE = 1
26465 @item SECTION_SCRIPT_ID_SCHEME_FILE = 3
26466 @item SECTION_SCRIPT_ID_PYTHON_TEXT = 4
26467 @item SECTION_SCRIPT_ID_SCHEME_TEXT = 6
26470 @subsubsection Script File Entries
26472 If the entry specifies a file, @value{GDBN} will look for the file first
26473 in the current directory and then along the source search path
26474 (@pxref{Source Path, ,Specifying Source Directories}),
26475 except that @file{$cdir} is not searched, since the compilation
26476 directory is not relevant to scripts.
26478 File entries can be placed in section @code{.debug_gdb_scripts} with,
26479 for example, this GCC macro for Python scripts.
26482 /* Note: The "MS" section flags are to remove duplicates. */
26483 #define DEFINE_GDB_PY_SCRIPT(script_name) \
26485 .pushsection \".debug_gdb_scripts\", \"MS\",@@progbits,1\n\
26486 .byte 1 /* Python */\n\
26487 .asciz \"" script_name "\"\n\
26493 For Guile scripts, replace @code{.byte 1} with @code{.byte 3}.
26494 Then one can reference the macro in a header or source file like this:
26497 DEFINE_GDB_PY_SCRIPT ("my-app-scripts.py")
26500 The script name may include directories if desired.
26502 Note that loading of this script file also requires accordingly configured
26503 @code{auto-load safe-path} (@pxref{Auto-loading safe path}).
26505 If the macro invocation is put in a header, any application or library
26506 using this header will get a reference to the specified script,
26507 and with the use of @code{"MS"} attributes on the section, the linker
26508 will remove duplicates.
26510 @subsubsection Script Text Entries
26512 Script text entries allow to put the executable script in the entry
26513 itself instead of loading it from a file.
26514 The first line of the entry, everything after the prefix byte and up to
26515 the first newline (@code{0xa}) character, is the script name, and must not
26516 contain any kind of space character, e.g., spaces or tabs.
26517 The rest of the entry, up to the trailing null byte, is the script to
26518 execute in the specified language. The name needs to be unique among
26519 all script names, as @value{GDBN} executes each script only once based
26522 Here is an example from file @file{py-section-script.c} in the @value{GDBN}
26526 #include "symcat.h"
26527 #include "gdb/section-scripts.h"
26529 ".pushsection \".debug_gdb_scripts\", \"MS\",@@progbits,1\n"
26530 ".byte " XSTRING (SECTION_SCRIPT_ID_PYTHON_TEXT) "\n"
26531 ".ascii \"gdb.inlined-script\\n\"\n"
26532 ".ascii \"class test_cmd (gdb.Command):\\n\"\n"
26533 ".ascii \" def __init__ (self):\\n\"\n"
26534 ".ascii \" super (test_cmd, self).__init__ ("
26535 "\\\"test-cmd\\\", gdb.COMMAND_OBSCURE)\\n\"\n"
26536 ".ascii \" def invoke (self, arg, from_tty):\\n\"\n"
26537 ".ascii \" print (\\\"test-cmd output, arg = %s\\\" % arg)\\n\"\n"
26538 ".ascii \"test_cmd ()\\n\"\n"
26544 Loading of inlined scripts requires a properly configured
26545 @code{auto-load safe-path} (@pxref{Auto-loading safe path}).
26546 The path to specify in @code{auto-load safe-path} is the path of the file
26547 containing the @code{.debug_gdb_scripts} section.
26549 @node Which flavor to choose?
26550 @subsection Which flavor to choose?
26552 Given the multiple ways of auto-loading extensions, it might not always
26553 be clear which one to choose. This section provides some guidance.
26556 Benefits of the @file{-gdb.@var{ext}} way:
26560 Can be used with file formats that don't support multiple sections.
26563 Ease of finding scripts for public libraries.
26565 Scripts specified in the @code{.debug_gdb_scripts} section are searched for
26566 in the source search path.
26567 For publicly installed libraries, e.g., @file{libstdc++}, there typically
26568 isn't a source directory in which to find the script.
26571 Doesn't require source code additions.
26575 Benefits of the @code{.debug_gdb_scripts} way:
26579 Works with static linking.
26581 Scripts for libraries done the @file{-gdb.@var{ext}} way require an objfile to
26582 trigger their loading. When an application is statically linked the only
26583 objfile available is the executable, and it is cumbersome to attach all the
26584 scripts from all the input libraries to the executable's
26585 @file{-gdb.@var{ext}} script.
26588 Works with classes that are entirely inlined.
26590 Some classes can be entirely inlined, and thus there may not be an associated
26591 shared library to attach a @file{-gdb.@var{ext}} script to.
26594 Scripts needn't be copied out of the source tree.
26596 In some circumstances, apps can be built out of large collections of internal
26597 libraries, and the build infrastructure necessary to install the
26598 @file{-gdb.@var{ext}} scripts in a place where @value{GDBN} can find them is
26599 cumbersome. It may be easier to specify the scripts in the
26600 @code{.debug_gdb_scripts} section as relative paths, and add a path to the
26601 top of the source tree to the source search path.
26604 @node Multiple Extension Languages
26605 @section Multiple Extension Languages
26607 The Guile and Python extension languages do not share any state,
26608 and generally do not interfere with each other.
26609 There are some things to be aware of, however.
26611 @subsection Python comes first
26613 Python was @value{GDBN}'s first extension language, and to avoid breaking
26614 existing behaviour Python comes first. This is generally solved by the
26615 ``first one wins'' principle. @value{GDBN} maintains a list of enabled
26616 extension languages, and when it makes a call to an extension language,
26617 (say to pretty-print a value), it tries each in turn until an extension
26618 language indicates it has performed the request (e.g., has returned the
26619 pretty-printed form of a value).
26620 This extends to errors while performing such requests: If an error happens
26621 while, for example, trying to pretty-print an object then the error is
26622 reported and any following extension languages are not tried.
26625 @section Creating new spellings of existing commands
26626 @cindex aliases for commands
26628 It is often useful to define alternate spellings of existing commands.
26629 For example, if a new @value{GDBN} command defined in Python has
26630 a long name to type, it is handy to have an abbreviated version of it
26631 that involves less typing.
26633 @value{GDBN} itself uses aliases. For example @samp{s} is an alias
26634 of the @samp{step} command even though it is otherwise an ambiguous
26635 abbreviation of other commands like @samp{set} and @samp{show}.
26637 Aliases are also used to provide shortened or more common versions
26638 of multi-word commands. For example, @value{GDBN} provides the
26639 @samp{tty} alias of the @samp{set inferior-tty} command.
26641 You can define a new alias with the @samp{alias} command.
26646 @item alias [-a] [--] @var{ALIAS} = @var{COMMAND}
26650 @var{ALIAS} specifies the name of the new alias.
26651 Each word of @var{ALIAS} must consist of letters, numbers, dashes and
26654 @var{COMMAND} specifies the name of an existing command
26655 that is being aliased.
26657 The @samp{-a} option specifies that the new alias is an abbreviation
26658 of the command. Abbreviations are not shown in command
26659 lists displayed by the @samp{help} command.
26661 The @samp{--} option specifies the end of options,
26662 and is useful when @var{ALIAS} begins with a dash.
26664 Here is a simple example showing how to make an abbreviation
26665 of a command so that there is less to type.
26666 Suppose you were tired of typing @samp{disas}, the current
26667 shortest unambiguous abbreviation of the @samp{disassemble} command
26668 and you wanted an even shorter version named @samp{di}.
26669 The following will accomplish this.
26672 (gdb) alias -a di = disas
26675 Note that aliases are different from user-defined commands.
26676 With a user-defined command, you also need to write documentation
26677 for it with the @samp{document} command.
26678 An alias automatically picks up the documentation of the existing command.
26680 Here is an example where we make @samp{elms} an abbreviation of
26681 @samp{elements} in the @samp{set print elements} command.
26682 This is to show that you can make an abbreviation of any part
26686 (gdb) alias -a set print elms = set print elements
26687 (gdb) alias -a show print elms = show print elements
26688 (gdb) set p elms 20
26690 Limit on string chars or array elements to print is 200.
26693 Note that if you are defining an alias of a @samp{set} command,
26694 and you want to have an alias for the corresponding @samp{show}
26695 command, then you need to define the latter separately.
26697 Unambiguously abbreviated commands are allowed in @var{COMMAND} and
26698 @var{ALIAS}, just as they are normally.
26701 (gdb) alias -a set pr elms = set p ele
26704 Finally, here is an example showing the creation of a one word
26705 alias for a more complex command.
26706 This creates alias @samp{spe} of the command @samp{set print elements}.
26709 (gdb) alias spe = set print elements
26714 @chapter Command Interpreters
26715 @cindex command interpreters
26717 @value{GDBN} supports multiple command interpreters, and some command
26718 infrastructure to allow users or user interface writers to switch
26719 between interpreters or run commands in other interpreters.
26721 @value{GDBN} currently supports two command interpreters, the console
26722 interpreter (sometimes called the command-line interpreter or @sc{cli})
26723 and the machine interface interpreter (or @sc{gdb/mi}). This manual
26724 describes both of these interfaces in great detail.
26726 By default, @value{GDBN} will start with the console interpreter.
26727 However, the user may choose to start @value{GDBN} with another
26728 interpreter by specifying the @option{-i} or @option{--interpreter}
26729 startup options. Defined interpreters include:
26733 @cindex console interpreter
26734 The traditional console or command-line interpreter. This is the most often
26735 used interpreter with @value{GDBN}. With no interpreter specified at runtime,
26736 @value{GDBN} will use this interpreter.
26739 @cindex mi interpreter
26740 The newest @sc{gdb/mi} interface (currently @code{mi3}). Used primarily
26741 by programs wishing to use @value{GDBN} as a backend for a debugger GUI
26742 or an IDE. For more information, see @ref{GDB/MI, ,The @sc{gdb/mi}
26746 @cindex mi3 interpreter
26747 The @sc{gdb/mi} interface introduced in @value{GDBN} 9.1.
26750 @cindex mi2 interpreter
26751 The @sc{gdb/mi} interface introduced in @value{GDBN} 6.0.
26754 @cindex mi1 interpreter
26755 The @sc{gdb/mi} interface introduced in @value{GDBN} 5.1.
26759 @cindex invoke another interpreter
26761 @kindex interpreter-exec
26762 You may execute commands in any interpreter from the current
26763 interpreter using the appropriate command. If you are running the
26764 console interpreter, simply use the @code{interpreter-exec} command:
26767 interpreter-exec mi "-data-list-register-names"
26770 @sc{gdb/mi} has a similar command, although it is only available in versions of
26771 @value{GDBN} which support @sc{gdb/mi} version 2 (or greater).
26773 Note that @code{interpreter-exec} only changes the interpreter for the
26774 duration of the specified command. It does not change the interpreter
26777 @cindex start a new independent interpreter
26779 Although you may only choose a single interpreter at startup, it is
26780 possible to run an independent interpreter on a specified input/output
26781 device (usually a tty).
26783 For example, consider a debugger GUI or IDE that wants to provide a
26784 @value{GDBN} console view. It may do so by embedding a terminal
26785 emulator widget in its GUI, starting @value{GDBN} in the traditional
26786 command-line mode with stdin/stdout/stderr redirected to that
26787 terminal, and then creating an MI interpreter running on a specified
26788 input/output device. The console interpreter created by @value{GDBN}
26789 at startup handles commands the user types in the terminal widget,
26790 while the GUI controls and synchronizes state with @value{GDBN} using
26791 the separate MI interpreter.
26793 To start a new secondary @dfn{user interface} running MI, use the
26794 @code{new-ui} command:
26797 @cindex new user interface
26799 new-ui @var{interpreter} @var{tty}
26802 The @var{interpreter} parameter specifies the interpreter to run.
26803 This accepts the same values as the @code{interpreter-exec} command.
26804 For example, @samp{console}, @samp{mi}, @samp{mi2}, etc. The
26805 @var{tty} parameter specifies the name of the bidirectional file the
26806 interpreter uses for input/output, usually the name of a
26807 pseudoterminal slave on Unix systems. For example:
26810 (@value{GDBP}) new-ui mi /dev/pts/9
26814 runs an MI interpreter on @file{/dev/pts/9}.
26817 @chapter @value{GDBN} Text User Interface
26819 @cindex Text User Interface
26822 * TUI Overview:: TUI overview
26823 * TUI Keys:: TUI key bindings
26824 * TUI Single Key Mode:: TUI single key mode
26825 * TUI Commands:: TUI-specific commands
26826 * TUI Configuration:: TUI configuration variables
26829 The @value{GDBN} Text User Interface (TUI) is a terminal
26830 interface which uses the @code{curses} library to show the source
26831 file, the assembly output, the program registers and @value{GDBN}
26832 commands in separate text windows. The TUI mode is supported only
26833 on platforms where a suitable version of the @code{curses} library
26836 The TUI mode is enabled by default when you invoke @value{GDBN} as
26837 @samp{@value{GDBP} -tui}.
26838 You can also switch in and out of TUI mode while @value{GDBN} runs by
26839 using various TUI commands and key bindings, such as @command{tui
26840 enable} or @kbd{C-x C-a}. @xref{TUI Commands, ,TUI Commands}, and
26841 @ref{TUI Keys, ,TUI Key Bindings}.
26844 @section TUI Overview
26846 In TUI mode, @value{GDBN} can display several text windows:
26850 This window is the @value{GDBN} command window with the @value{GDBN}
26851 prompt and the @value{GDBN} output. The @value{GDBN} input is still
26852 managed using readline.
26855 The source window shows the source file of the program. The current
26856 line and active breakpoints are displayed in this window.
26859 The assembly window shows the disassembly output of the program.
26862 This window shows the processor registers. Registers are highlighted
26863 when their values change.
26866 The source and assembly windows show the current program position
26867 by highlighting the current line and marking it with a @samp{>} marker.
26868 Breakpoints are indicated with two markers. The first marker
26869 indicates the breakpoint type:
26873 Breakpoint which was hit at least once.
26876 Breakpoint which was never hit.
26879 Hardware breakpoint which was hit at least once.
26882 Hardware breakpoint which was never hit.
26885 The second marker indicates whether the breakpoint is enabled or not:
26889 Breakpoint is enabled.
26892 Breakpoint is disabled.
26895 The source, assembly and register windows are updated when the current
26896 thread changes, when the frame changes, or when the program counter
26899 These windows are not all visible at the same time. The command
26900 window is always visible. The others can be arranged in several
26911 source and assembly,
26914 source and registers, or
26917 assembly and registers.
26920 A status line above the command window shows the following information:
26924 Indicates the current @value{GDBN} target.
26925 (@pxref{Targets, ,Specifying a Debugging Target}).
26928 Gives the current process or thread number.
26929 When no process is being debugged, this field is set to @code{No process}.
26932 Gives the current function name for the selected frame.
26933 The name is demangled if demangling is turned on (@pxref{Print Settings}).
26934 When there is no symbol corresponding to the current program counter,
26935 the string @code{??} is displayed.
26938 Indicates the current line number for the selected frame.
26939 When the current line number is not known, the string @code{??} is displayed.
26942 Indicates the current program counter address.
26946 @section TUI Key Bindings
26947 @cindex TUI key bindings
26949 The TUI installs several key bindings in the readline keymaps
26950 @ifset SYSTEM_READLINE
26951 (@pxref{Command Line Editing, , , rluserman, GNU Readline Library}).
26953 @ifclear SYSTEM_READLINE
26954 (@pxref{Command Line Editing}).
26956 The following key bindings are installed for both TUI mode and the
26957 @value{GDBN} standard mode.
26966 Enter or leave the TUI mode. When leaving the TUI mode,
26967 the curses window management stops and @value{GDBN} operates using
26968 its standard mode, writing on the terminal directly. When reentering
26969 the TUI mode, control is given back to the curses windows.
26970 The screen is then refreshed.
26974 Use a TUI layout with only one window. The layout will
26975 either be @samp{source} or @samp{assembly}. When the TUI mode
26976 is not active, it will switch to the TUI mode.
26978 Think of this key binding as the Emacs @kbd{C-x 1} binding.
26982 Use a TUI layout with at least two windows. When the current
26983 layout already has two windows, the next layout with two windows is used.
26984 When a new layout is chosen, one window will always be common to the
26985 previous layout and the new one.
26987 Think of it as the Emacs @kbd{C-x 2} binding.
26991 Change the active window. The TUI associates several key bindings
26992 (like scrolling and arrow keys) with the active window. This command
26993 gives the focus to the next TUI window.
26995 Think of it as the Emacs @kbd{C-x o} binding.
26999 Switch in and out of the TUI SingleKey mode that binds single
27000 keys to @value{GDBN} commands (@pxref{TUI Single Key Mode}).
27003 The following key bindings only work in the TUI mode:
27008 Scroll the active window one page up.
27012 Scroll the active window one page down.
27016 Scroll the active window one line up.
27020 Scroll the active window one line down.
27024 Scroll the active window one column left.
27028 Scroll the active window one column right.
27032 Refresh the screen.
27035 Because the arrow keys scroll the active window in the TUI mode, they
27036 are not available for their normal use by readline unless the command
27037 window has the focus. When another window is active, you must use
27038 other readline key bindings such as @kbd{C-p}, @kbd{C-n}, @kbd{C-b}
27039 and @kbd{C-f} to control the command window.
27041 @node TUI Single Key Mode
27042 @section TUI Single Key Mode
27043 @cindex TUI single key mode
27045 The TUI also provides a @dfn{SingleKey} mode, which binds several
27046 frequently used @value{GDBN} commands to single keys. Type @kbd{C-x s} to
27047 switch into this mode, where the following key bindings are used:
27050 @kindex c @r{(SingleKey TUI key)}
27054 @kindex d @r{(SingleKey TUI key)}
27058 @kindex f @r{(SingleKey TUI key)}
27062 @kindex n @r{(SingleKey TUI key)}
27066 @kindex o @r{(SingleKey TUI key)}
27068 nexti. The shortcut letter @samp{o} stands for ``step Over''.
27070 @kindex q @r{(SingleKey TUI key)}
27072 exit the SingleKey mode.
27074 @kindex r @r{(SingleKey TUI key)}
27078 @kindex s @r{(SingleKey TUI key)}
27082 @kindex i @r{(SingleKey TUI key)}
27084 stepi. The shortcut letter @samp{i} stands for ``step Into''.
27086 @kindex u @r{(SingleKey TUI key)}
27090 @kindex v @r{(SingleKey TUI key)}
27094 @kindex w @r{(SingleKey TUI key)}
27099 Other keys temporarily switch to the @value{GDBN} command prompt.
27100 The key that was pressed is inserted in the editing buffer so that
27101 it is possible to type most @value{GDBN} commands without interaction
27102 with the TUI SingleKey mode. Once the command is entered the TUI
27103 SingleKey mode is restored. The only way to permanently leave
27104 this mode is by typing @kbd{q} or @kbd{C-x s}.
27108 @section TUI-specific Commands
27109 @cindex TUI commands
27111 The TUI has specific commands to control the text windows.
27112 These commands are always available, even when @value{GDBN} is not in
27113 the TUI mode. When @value{GDBN} is in the standard mode, most
27114 of these commands will automatically switch to the TUI mode.
27116 Note that if @value{GDBN}'s @code{stdout} is not connected to a
27117 terminal, or @value{GDBN} has been started with the machine interface
27118 interpreter (@pxref{GDB/MI, ,The @sc{gdb/mi} Interface}), most of
27119 these commands will fail with an error, because it would not be
27120 possible or desirable to enable curses window management.
27125 Activate TUI mode. The last active TUI window layout will be used if
27126 TUI mode has prevsiouly been used in the current debugging session,
27127 otherwise a default layout is used.
27130 @kindex tui disable
27131 Disable TUI mode, returning to the console interpreter.
27135 List and give the size of all displayed windows.
27137 @item layout @var{name}
27139 Changes which TUI windows are displayed. In each layout the command
27140 window is always displayed, the @var{name} parameter controls which
27141 additional windows are displayed, and can be any of the following:
27145 Display the next layout.
27148 Display the previous layout.
27151 Display the source and command windows.
27154 Display the assembly and command windows.
27157 Display the source, assembly, and command windows.
27160 When in @code{src} layout display the register, source, and command
27161 windows. When in @code{asm} or @code{split} layout display the
27162 register, assembler, and command windows.
27165 @item focus @var{name}
27167 Changes which TUI window is currently active for scrolling. The
27168 @var{name} parameter can be any of the following:
27172 Make the next window active for scrolling.
27175 Make the previous window active for scrolling.
27178 Make the source window active for scrolling.
27181 Make the assembly window active for scrolling.
27184 Make the register window active for scrolling.
27187 Make the command window active for scrolling.
27192 Refresh the screen. This is similar to typing @kbd{C-L}.
27194 @item tui reg @var{group}
27196 Changes the register group displayed in the tui register window to
27197 @var{group}. If the register window is not currently displayed this
27198 command will cause the register window to be displayed. The list of
27199 register groups, as well as their order is target specific. The
27200 following groups are available on most targets:
27203 Repeatedly selecting this group will cause the display to cycle
27204 through all of the available register groups.
27207 Repeatedly selecting this group will cause the display to cycle
27208 through all of the available register groups in the reverse order to
27212 Display the general registers.
27214 Display the floating point registers.
27216 Display the system registers.
27218 Display the vector registers.
27220 Display all registers.
27225 Update the source window and the current execution point.
27227 @item winheight @var{name} +@var{count}
27228 @itemx winheight @var{name} -@var{count}
27230 Change the height of the window @var{name} by @var{count}
27231 lines. Positive counts increase the height, while negative counts
27232 decrease it. The @var{name} parameter can be one of @code{src} (the
27233 source window), @code{cmd} (the command window), @code{asm} (the
27234 disassembly window), or @code{regs} (the register display window).
27237 @node TUI Configuration
27238 @section TUI Configuration Variables
27239 @cindex TUI configuration variables
27241 Several configuration variables control the appearance of TUI windows.
27244 @item set tui border-kind @var{kind}
27245 @kindex set tui border-kind
27246 Select the border appearance for the source, assembly and register windows.
27247 The possible values are the following:
27250 Use a space character to draw the border.
27253 Use @sc{ascii} characters @samp{+}, @samp{-} and @samp{|} to draw the border.
27256 Use the Alternate Character Set to draw the border. The border is
27257 drawn using character line graphics if the terminal supports them.
27260 @item set tui border-mode @var{mode}
27261 @kindex set tui border-mode
27262 @itemx set tui active-border-mode @var{mode}
27263 @kindex set tui active-border-mode
27264 Select the display attributes for the borders of the inactive windows
27265 or the active window. The @var{mode} can be one of the following:
27268 Use normal attributes to display the border.
27274 Use reverse video mode.
27277 Use half bright mode.
27279 @item half-standout
27280 Use half bright and standout mode.
27283 Use extra bright or bold mode.
27285 @item bold-standout
27286 Use extra bright or bold and standout mode.
27289 @item set tui tab-width @var{nchars}
27290 @kindex set tui tab-width
27292 Set the width of tab stops to be @var{nchars} characters. This
27293 setting affects the display of TAB characters in the source and
27298 @chapter Using @value{GDBN} under @sc{gnu} Emacs
27301 @cindex @sc{gnu} Emacs
27302 A special interface allows you to use @sc{gnu} Emacs to view (and
27303 edit) the source files for the program you are debugging with
27306 To use this interface, use the command @kbd{M-x gdb} in Emacs. Give the
27307 executable file you want to debug as an argument. This command starts
27308 @value{GDBN} as a subprocess of Emacs, with input and output through a newly
27309 created Emacs buffer.
27310 @c (Do not use the @code{-tui} option to run @value{GDBN} from Emacs.)
27312 Running @value{GDBN} under Emacs can be just like running @value{GDBN} normally except for two
27317 All ``terminal'' input and output goes through an Emacs buffer, called
27320 This applies both to @value{GDBN} commands and their output, and to the input
27321 and output done by the program you are debugging.
27323 This is useful because it means that you can copy the text of previous
27324 commands and input them again; you can even use parts of the output
27327 All the facilities of Emacs' Shell mode are available for interacting
27328 with your program. In particular, you can send signals the usual
27329 way---for example, @kbd{C-c C-c} for an interrupt, @kbd{C-c C-z} for a
27333 @value{GDBN} displays source code through Emacs.
27335 Each time @value{GDBN} displays a stack frame, Emacs automatically finds the
27336 source file for that frame and puts an arrow (@samp{=>}) at the
27337 left margin of the current line. Emacs uses a separate buffer for
27338 source display, and splits the screen to show both your @value{GDBN} session
27341 Explicit @value{GDBN} @code{list} or search commands still produce output as
27342 usual, but you probably have no reason to use them from Emacs.
27345 We call this @dfn{text command mode}. Emacs 22.1, and later, also uses
27346 a graphical mode, enabled by default, which provides further buffers
27347 that can control the execution and describe the state of your program.
27348 @xref{GDB Graphical Interface,,, Emacs, The @sc{gnu} Emacs Manual}.
27350 If you specify an absolute file name when prompted for the @kbd{M-x
27351 gdb} argument, then Emacs sets your current working directory to where
27352 your program resides. If you only specify the file name, then Emacs
27353 sets your current working directory to the directory associated
27354 with the previous buffer. In this case, @value{GDBN} may find your
27355 program by searching your environment's @code{PATH} variable, but on
27356 some operating systems it might not find the source. So, although the
27357 @value{GDBN} input and output session proceeds normally, the auxiliary
27358 buffer does not display the current source and line of execution.
27360 The initial working directory of @value{GDBN} is printed on the top
27361 line of the GUD buffer and this serves as a default for the commands
27362 that specify files for @value{GDBN} to operate on. @xref{Files,
27363 ,Commands to Specify Files}.
27365 By default, @kbd{M-x gdb} calls the program called @file{gdb}. If you
27366 need to call @value{GDBN} by a different name (for example, if you
27367 keep several configurations around, with different names) you can
27368 customize the Emacs variable @code{gud-gdb-command-name} to run the
27371 In the GUD buffer, you can use these special Emacs commands in
27372 addition to the standard Shell mode commands:
27376 Describe the features of Emacs' GUD Mode.
27379 Execute to another source line, like the @value{GDBN} @code{step} command; also
27380 update the display window to show the current file and location.
27383 Execute to next source line in this function, skipping all function
27384 calls, like the @value{GDBN} @code{next} command. Then update the display window
27385 to show the current file and location.
27388 Execute one instruction, like the @value{GDBN} @code{stepi} command; update
27389 display window accordingly.
27392 Execute until exit from the selected stack frame, like the @value{GDBN}
27393 @code{finish} command.
27396 Continue execution of your program, like the @value{GDBN} @code{continue}
27400 Go up the number of frames indicated by the numeric argument
27401 (@pxref{Arguments, , Numeric Arguments, Emacs, The @sc{gnu} Emacs Manual}),
27402 like the @value{GDBN} @code{up} command.
27405 Go down the number of frames indicated by the numeric argument, like the
27406 @value{GDBN} @code{down} command.
27409 In any source file, the Emacs command @kbd{C-x @key{SPC}} (@code{gud-break})
27410 tells @value{GDBN} to set a breakpoint on the source line point is on.
27412 In text command mode, if you type @kbd{M-x speedbar}, Emacs displays a
27413 separate frame which shows a backtrace when the GUD buffer is current.
27414 Move point to any frame in the stack and type @key{RET} to make it
27415 become the current frame and display the associated source in the
27416 source buffer. Alternatively, click @kbd{Mouse-2} to make the
27417 selected frame become the current one. In graphical mode, the
27418 speedbar displays watch expressions.
27420 If you accidentally delete the source-display buffer, an easy way to get
27421 it back is to type the command @code{f} in the @value{GDBN} buffer, to
27422 request a frame display; when you run under Emacs, this recreates
27423 the source buffer if necessary to show you the context of the current
27426 The source files displayed in Emacs are in ordinary Emacs buffers
27427 which are visiting the source files in the usual way. You can edit
27428 the files with these buffers if you wish; but keep in mind that @value{GDBN}
27429 communicates with Emacs in terms of line numbers. If you add or
27430 delete lines from the text, the line numbers that @value{GDBN} knows cease
27431 to correspond properly with the code.
27433 A more detailed description of Emacs' interaction with @value{GDBN} is
27434 given in the Emacs manual (@pxref{Debuggers,,, Emacs, The @sc{gnu}
27438 @chapter The @sc{gdb/mi} Interface
27440 @unnumberedsec Function and Purpose
27442 @cindex @sc{gdb/mi}, its purpose
27443 @sc{gdb/mi} is a line based machine oriented text interface to
27444 @value{GDBN} and is activated by specifying using the
27445 @option{--interpreter} command line option (@pxref{Mode Options}). It
27446 is specifically intended to support the development of systems which
27447 use the debugger as just one small component of a larger system.
27449 This chapter is a specification of the @sc{gdb/mi} interface. It is written
27450 in the form of a reference manual.
27452 Note that @sc{gdb/mi} is still under construction, so some of the
27453 features described below are incomplete and subject to change
27454 (@pxref{GDB/MI Development and Front Ends, , @sc{gdb/mi} Development and Front Ends}).
27456 @unnumberedsec Notation and Terminology
27458 @cindex notational conventions, for @sc{gdb/mi}
27459 This chapter uses the following notation:
27463 @code{|} separates two alternatives.
27466 @code{[ @var{something} ]} indicates that @var{something} is optional:
27467 it may or may not be given.
27470 @code{( @var{group} )*} means that @var{group} inside the parentheses
27471 may repeat zero or more times.
27474 @code{( @var{group} )+} means that @var{group} inside the parentheses
27475 may repeat one or more times.
27478 @code{"@var{string}"} means a literal @var{string}.
27482 @heading Dependencies
27486 * GDB/MI General Design::
27487 * GDB/MI Command Syntax::
27488 * GDB/MI Compatibility with CLI::
27489 * GDB/MI Development and Front Ends::
27490 * GDB/MI Output Records::
27491 * GDB/MI Simple Examples::
27492 * GDB/MI Command Description Format::
27493 * GDB/MI Breakpoint Commands::
27494 * GDB/MI Catchpoint Commands::
27495 * GDB/MI Program Context::
27496 * GDB/MI Thread Commands::
27497 * GDB/MI Ada Tasking Commands::
27498 * GDB/MI Program Execution::
27499 * GDB/MI Stack Manipulation::
27500 * GDB/MI Variable Objects::
27501 * GDB/MI Data Manipulation::
27502 * GDB/MI Tracepoint Commands::
27503 * GDB/MI Symbol Query::
27504 * GDB/MI File Commands::
27506 * GDB/MI Kod Commands::
27507 * GDB/MI Memory Overlay Commands::
27508 * GDB/MI Signal Handling Commands::
27510 * GDB/MI Target Manipulation::
27511 * GDB/MI File Transfer Commands::
27512 * GDB/MI Ada Exceptions Commands::
27513 * GDB/MI Support Commands::
27514 * GDB/MI Miscellaneous Commands::
27517 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
27518 @node GDB/MI General Design
27519 @section @sc{gdb/mi} General Design
27520 @cindex GDB/MI General Design
27522 Interaction of a @sc{GDB/MI} frontend with @value{GDBN} involves three
27523 parts---commands sent to @value{GDBN}, responses to those commands
27524 and notifications. Each command results in exactly one response,
27525 indicating either successful completion of the command, or an error.
27526 For the commands that do not resume the target, the response contains the
27527 requested information. For the commands that resume the target, the
27528 response only indicates whether the target was successfully resumed.
27529 Notifications is the mechanism for reporting changes in the state of the
27530 target, or in @value{GDBN} state, that cannot conveniently be associated with
27531 a command and reported as part of that command response.
27533 The important examples of notifications are:
27537 Exec notifications. These are used to report changes in
27538 target state---when a target is resumed, or stopped. It would not
27539 be feasible to include this information in response of resuming
27540 commands, because one resume commands can result in multiple events in
27541 different threads. Also, quite some time may pass before any event
27542 happens in the target, while a frontend needs to know whether the resuming
27543 command itself was successfully executed.
27546 Console output, and status notifications. Console output
27547 notifications are used to report output of CLI commands, as well as
27548 diagnostics for other commands. Status notifications are used to
27549 report the progress of a long-running operation. Naturally, including
27550 this information in command response would mean no output is produced
27551 until the command is finished, which is undesirable.
27554 General notifications. Commands may have various side effects on
27555 the @value{GDBN} or target state beyond their official purpose. For example,
27556 a command may change the selected thread. Although such changes can
27557 be included in command response, using notification allows for more
27558 orthogonal frontend design.
27562 There's no guarantee that whenever an MI command reports an error,
27563 @value{GDBN} or the target are in any specific state, and especially,
27564 the state is not reverted to the state before the MI command was
27565 processed. Therefore, whenever an MI command results in an error,
27566 we recommend that the frontend refreshes all the information shown in
27567 the user interface.
27571 * Context management::
27572 * Asynchronous and non-stop modes::
27576 @node Context management
27577 @subsection Context management
27579 @subsubsection Threads and Frames
27581 In most cases when @value{GDBN} accesses the target, this access is
27582 done in context of a specific thread and frame (@pxref{Frames}).
27583 Often, even when accessing global data, the target requires that a thread
27584 be specified. The CLI interface maintains the selected thread and frame,
27585 and supplies them to target on each command. This is convenient,
27586 because a command line user would not want to specify that information
27587 explicitly on each command, and because user interacts with
27588 @value{GDBN} via a single terminal, so no confusion is possible as
27589 to what thread and frame are the current ones.
27591 In the case of MI, the concept of selected thread and frame is less
27592 useful. First, a frontend can easily remember this information
27593 itself. Second, a graphical frontend can have more than one window,
27594 each one used for debugging a different thread, and the frontend might
27595 want to access additional threads for internal purposes. This
27596 increases the risk that by relying on implicitly selected thread, the
27597 frontend may be operating on a wrong one. Therefore, each MI command
27598 should explicitly specify which thread and frame to operate on. To
27599 make it possible, each MI command accepts the @samp{--thread} and
27600 @samp{--frame} options, the value to each is @value{GDBN} global
27601 identifier for thread and frame to operate on.
27603 Usually, each top-level window in a frontend allows the user to select
27604 a thread and a frame, and remembers the user selection for further
27605 operations. However, in some cases @value{GDBN} may suggest that the
27606 current thread or frame be changed. For example, when stopping on a
27607 breakpoint it is reasonable to switch to the thread where breakpoint is
27608 hit. For another example, if the user issues the CLI @samp{thread} or
27609 @samp{frame} commands via the frontend, it is desirable to change the
27610 frontend's selection to the one specified by user. @value{GDBN}
27611 communicates the suggestion to change current thread and frame using the
27612 @samp{=thread-selected} notification.
27614 Note that historically, MI shares the selected thread with CLI, so
27615 frontends used the @code{-thread-select} to execute commands in the
27616 right context. However, getting this to work right is cumbersome. The
27617 simplest way is for frontend to emit @code{-thread-select} command
27618 before every command. This doubles the number of commands that need
27619 to be sent. The alternative approach is to suppress @code{-thread-select}
27620 if the selected thread in @value{GDBN} is supposed to be identical to the
27621 thread the frontend wants to operate on. However, getting this
27622 optimization right can be tricky. In particular, if the frontend
27623 sends several commands to @value{GDBN}, and one of the commands changes the
27624 selected thread, then the behaviour of subsequent commands will
27625 change. So, a frontend should either wait for response from such
27626 problematic commands, or explicitly add @code{-thread-select} for
27627 all subsequent commands. No frontend is known to do this exactly
27628 right, so it is suggested to just always pass the @samp{--thread} and
27629 @samp{--frame} options.
27631 @subsubsection Language
27633 The execution of several commands depends on which language is selected.
27634 By default, the current language (@pxref{show language}) is used.
27635 But for commands known to be language-sensitive, it is recommended
27636 to use the @samp{--language} option. This option takes one argument,
27637 which is the name of the language to use while executing the command.
27641 -data-evaluate-expression --language c "sizeof (void*)"
27646 The valid language names are the same names accepted by the
27647 @samp{set language} command (@pxref{Manually}), excluding @samp{auto},
27648 @samp{local} or @samp{unknown}.
27650 @node Asynchronous and non-stop modes
27651 @subsection Asynchronous command execution and non-stop mode
27653 On some targets, @value{GDBN} is capable of processing MI commands
27654 even while the target is running. This is called @dfn{asynchronous
27655 command execution} (@pxref{Background Execution}). The frontend may
27656 specify a preferrence for asynchronous execution using the
27657 @code{-gdb-set mi-async 1} command, which should be emitted before
27658 either running the executable or attaching to the target. After the
27659 frontend has started the executable or attached to the target, it can
27660 find if asynchronous execution is enabled using the
27661 @code{-list-target-features} command.
27664 @item -gdb-set mi-async on
27665 @item -gdb-set mi-async off
27666 Set whether MI is in asynchronous mode.
27668 When @code{off}, which is the default, MI execution commands (e.g.,
27669 @code{-exec-continue}) are foreground commands, and @value{GDBN} waits
27670 for the program to stop before processing further commands.
27672 When @code{on}, MI execution commands are background execution
27673 commands (e.g., @code{-exec-continue} becomes the equivalent of the
27674 @code{c&} CLI command), and so @value{GDBN} is capable of processing
27675 MI commands even while the target is running.
27677 @item -gdb-show mi-async
27678 Show whether MI asynchronous mode is enabled.
27681 Note: In @value{GDBN} version 7.7 and earlier, this option was called
27682 @code{target-async} instead of @code{mi-async}, and it had the effect
27683 of both putting MI in asynchronous mode and making CLI background
27684 commands possible. CLI background commands are now always possible
27685 ``out of the box'' if the target supports them. The old spelling is
27686 kept as a deprecated alias for backwards compatibility.
27688 Even if @value{GDBN} can accept a command while target is running,
27689 many commands that access the target do not work when the target is
27690 running. Therefore, asynchronous command execution is most useful
27691 when combined with non-stop mode (@pxref{Non-Stop Mode}). Then,
27692 it is possible to examine the state of one thread, while other threads
27695 When a given thread is running, MI commands that try to access the
27696 target in the context of that thread may not work, or may work only on
27697 some targets. In particular, commands that try to operate on thread's
27698 stack will not work, on any target. Commands that read memory, or
27699 modify breakpoints, may work or not work, depending on the target. Note
27700 that even commands that operate on global state, such as @code{print},
27701 @code{set}, and breakpoint commands, still access the target in the
27702 context of a specific thread, so frontend should try to find a
27703 stopped thread and perform the operation on that thread (using the
27704 @samp{--thread} option).
27706 Which commands will work in the context of a running thread is
27707 highly target dependent. However, the two commands
27708 @code{-exec-interrupt}, to stop a thread, and @code{-thread-info},
27709 to find the state of a thread, will always work.
27711 @node Thread groups
27712 @subsection Thread groups
27713 @value{GDBN} may be used to debug several processes at the same time.
27714 On some platfroms, @value{GDBN} may support debugging of several
27715 hardware systems, each one having several cores with several different
27716 processes running on each core. This section describes the MI
27717 mechanism to support such debugging scenarios.
27719 The key observation is that regardless of the structure of the
27720 target, MI can have a global list of threads, because most commands that
27721 accept the @samp{--thread} option do not need to know what process that
27722 thread belongs to. Therefore, it is not necessary to introduce
27723 neither additional @samp{--process} option, nor an notion of the
27724 current process in the MI interface. The only strictly new feature
27725 that is required is the ability to find how the threads are grouped
27728 To allow the user to discover such grouping, and to support arbitrary
27729 hierarchy of machines/cores/processes, MI introduces the concept of a
27730 @dfn{thread group}. Thread group is a collection of threads and other
27731 thread groups. A thread group always has a string identifier, a type,
27732 and may have additional attributes specific to the type. A new
27733 command, @code{-list-thread-groups}, returns the list of top-level
27734 thread groups, which correspond to processes that @value{GDBN} is
27735 debugging at the moment. By passing an identifier of a thread group
27736 to the @code{-list-thread-groups} command, it is possible to obtain
27737 the members of specific thread group.
27739 To allow the user to easily discover processes, and other objects, he
27740 wishes to debug, a concept of @dfn{available thread group} is
27741 introduced. Available thread group is an thread group that
27742 @value{GDBN} is not debugging, but that can be attached to, using the
27743 @code{-target-attach} command. The list of available top-level thread
27744 groups can be obtained using @samp{-list-thread-groups --available}.
27745 In general, the content of a thread group may be only retrieved only
27746 after attaching to that thread group.
27748 Thread groups are related to inferiors (@pxref{Inferiors and
27749 Programs}). Each inferior corresponds to a thread group of a special
27750 type @samp{process}, and some additional operations are permitted on
27751 such thread groups.
27753 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
27754 @node GDB/MI Command Syntax
27755 @section @sc{gdb/mi} Command Syntax
27758 * GDB/MI Input Syntax::
27759 * GDB/MI Output Syntax::
27762 @node GDB/MI Input Syntax
27763 @subsection @sc{gdb/mi} Input Syntax
27765 @cindex input syntax for @sc{gdb/mi}
27766 @cindex @sc{gdb/mi}, input syntax
27768 @item @var{command} @expansion{}
27769 @code{@var{cli-command} | @var{mi-command}}
27771 @item @var{cli-command} @expansion{}
27772 @code{[ @var{token} ] @var{cli-command} @var{nl}}, where
27773 @var{cli-command} is any existing @value{GDBN} CLI command.
27775 @item @var{mi-command} @expansion{}
27776 @code{[ @var{token} ] "-" @var{operation} ( " " @var{option} )*
27777 @code{[} " --" @code{]} ( " " @var{parameter} )* @var{nl}}
27779 @item @var{token} @expansion{}
27780 "any sequence of digits"
27782 @item @var{option} @expansion{}
27783 @code{"-" @var{parameter} [ " " @var{parameter} ]}
27785 @item @var{parameter} @expansion{}
27786 @code{@var{non-blank-sequence} | @var{c-string}}
27788 @item @var{operation} @expansion{}
27789 @emph{any of the operations described in this chapter}
27791 @item @var{non-blank-sequence} @expansion{}
27792 @emph{anything, provided it doesn't contain special characters such as
27793 "-", @var{nl}, """ and of course " "}
27795 @item @var{c-string} @expansion{}
27796 @code{""" @var{seven-bit-iso-c-string-content} """}
27798 @item @var{nl} @expansion{}
27807 The CLI commands are still handled by the @sc{mi} interpreter; their
27808 output is described below.
27811 The @code{@var{token}}, when present, is passed back when the command
27815 Some @sc{mi} commands accept optional arguments as part of the parameter
27816 list. Each option is identified by a leading @samp{-} (dash) and may be
27817 followed by an optional argument parameter. Options occur first in the
27818 parameter list and can be delimited from normal parameters using
27819 @samp{--} (this is useful when some parameters begin with a dash).
27826 We want easy access to the existing CLI syntax (for debugging).
27829 We want it to be easy to spot a @sc{mi} operation.
27832 @node GDB/MI Output Syntax
27833 @subsection @sc{gdb/mi} Output Syntax
27835 @cindex output syntax of @sc{gdb/mi}
27836 @cindex @sc{gdb/mi}, output syntax
27837 The output from @sc{gdb/mi} consists of zero or more out-of-band records
27838 followed, optionally, by a single result record. This result record
27839 is for the most recent command. The sequence of output records is
27840 terminated by @samp{(gdb)}.
27842 If an input command was prefixed with a @code{@var{token}} then the
27843 corresponding output for that command will also be prefixed by that same
27847 @item @var{output} @expansion{}
27848 @code{( @var{out-of-band-record} )* [ @var{result-record} ] "(gdb)" @var{nl}}
27850 @item @var{result-record} @expansion{}
27851 @code{ [ @var{token} ] "^" @var{result-class} ( "," @var{result} )* @var{nl}}
27853 @item @var{out-of-band-record} @expansion{}
27854 @code{@var{async-record} | @var{stream-record}}
27856 @item @var{async-record} @expansion{}
27857 @code{@var{exec-async-output} | @var{status-async-output} | @var{notify-async-output}}
27859 @item @var{exec-async-output} @expansion{}
27860 @code{[ @var{token} ] "*" @var{async-output nl}}
27862 @item @var{status-async-output} @expansion{}
27863 @code{[ @var{token} ] "+" @var{async-output nl}}
27865 @item @var{notify-async-output} @expansion{}
27866 @code{[ @var{token} ] "=" @var{async-output nl}}
27868 @item @var{async-output} @expansion{}
27869 @code{@var{async-class} ( "," @var{result} )*}
27871 @item @var{result-class} @expansion{}
27872 @code{"done" | "running" | "connected" | "error" | "exit"}
27874 @item @var{async-class} @expansion{}
27875 @code{"stopped" | @var{others}} (where @var{others} will be added
27876 depending on the needs---this is still in development).
27878 @item @var{result} @expansion{}
27879 @code{ @var{variable} "=" @var{value}}
27881 @item @var{variable} @expansion{}
27882 @code{ @var{string} }
27884 @item @var{value} @expansion{}
27885 @code{ @var{const} | @var{tuple} | @var{list} }
27887 @item @var{const} @expansion{}
27888 @code{@var{c-string}}
27890 @item @var{tuple} @expansion{}
27891 @code{ "@{@}" | "@{" @var{result} ( "," @var{result} )* "@}" }
27893 @item @var{list} @expansion{}
27894 @code{ "[]" | "[" @var{value} ( "," @var{value} )* "]" | "["
27895 @var{result} ( "," @var{result} )* "]" }
27897 @item @var{stream-record} @expansion{}
27898 @code{@var{console-stream-output} | @var{target-stream-output} | @var{log-stream-output}}
27900 @item @var{console-stream-output} @expansion{}
27901 @code{"~" @var{c-string nl}}
27903 @item @var{target-stream-output} @expansion{}
27904 @code{"@@" @var{c-string nl}}
27906 @item @var{log-stream-output} @expansion{}
27907 @code{"&" @var{c-string nl}}
27909 @item @var{nl} @expansion{}
27912 @item @var{token} @expansion{}
27913 @emph{any sequence of digits}.
27921 All output sequences end in a single line containing a period.
27924 The @code{@var{token}} is from the corresponding request. Note that
27925 for all async output, while the token is allowed by the grammar and
27926 may be output by future versions of @value{GDBN} for select async
27927 output messages, it is generally omitted. Frontends should treat
27928 all async output as reporting general changes in the state of the
27929 target and there should be no need to associate async output to any
27933 @cindex status output in @sc{gdb/mi}
27934 @var{status-async-output} contains on-going status information about the
27935 progress of a slow operation. It can be discarded. All status output is
27936 prefixed by @samp{+}.
27939 @cindex async output in @sc{gdb/mi}
27940 @var{exec-async-output} contains asynchronous state change on the target
27941 (stopped, started, disappeared). All async output is prefixed by
27945 @cindex notify output in @sc{gdb/mi}
27946 @var{notify-async-output} contains supplementary information that the
27947 client should handle (e.g., a new breakpoint information). All notify
27948 output is prefixed by @samp{=}.
27951 @cindex console output in @sc{gdb/mi}
27952 @var{console-stream-output} is output that should be displayed as is in the
27953 console. It is the textual response to a CLI command. All the console
27954 output is prefixed by @samp{~}.
27957 @cindex target output in @sc{gdb/mi}
27958 @var{target-stream-output} is the output produced by the target program.
27959 All the target output is prefixed by @samp{@@}.
27962 @cindex log output in @sc{gdb/mi}
27963 @var{log-stream-output} is output text coming from @value{GDBN}'s internals, for
27964 instance messages that should be displayed as part of an error log. All
27965 the log output is prefixed by @samp{&}.
27968 @cindex list output in @sc{gdb/mi}
27969 New @sc{gdb/mi} commands should only output @var{lists} containing
27975 @xref{GDB/MI Stream Records, , @sc{gdb/mi} Stream Records}, for more
27976 details about the various output records.
27978 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
27979 @node GDB/MI Compatibility with CLI
27980 @section @sc{gdb/mi} Compatibility with CLI
27982 @cindex compatibility, @sc{gdb/mi} and CLI
27983 @cindex @sc{gdb/mi}, compatibility with CLI
27985 For the developers convenience CLI commands can be entered directly,
27986 but there may be some unexpected behaviour. For example, commands
27987 that query the user will behave as if the user replied yes, breakpoint
27988 command lists are not executed and some CLI commands, such as
27989 @code{if}, @code{when} and @code{define}, prompt for further input with
27990 @samp{>}, which is not valid MI output.
27992 This feature may be removed at some stage in the future and it is
27993 recommended that front ends use the @code{-interpreter-exec} command
27994 (@pxref{-interpreter-exec}).
27996 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
27997 @node GDB/MI Development and Front Ends
27998 @section @sc{gdb/mi} Development and Front Ends
27999 @cindex @sc{gdb/mi} development
28001 The application which takes the MI output and presents the state of the
28002 program being debugged to the user is called a @dfn{front end}.
28004 Since @sc{gdb/mi} is used by a variety of front ends to @value{GDBN}, changes
28005 to the MI interface may break existing usage. This section describes how the
28006 protocol changes and how to request previous version of the protocol when it
28009 Some changes in MI need not break a carefully designed front end, and
28010 for these the MI version will remain unchanged. The following is a
28011 list of changes that may occur within one level, so front ends should
28012 parse MI output in a way that can handle them:
28016 New MI commands may be added.
28019 New fields may be added to the output of any MI command.
28022 The range of values for fields with specified values, e.g.,
28023 @code{in_scope} (@pxref{-var-update}) may be extended.
28025 @c The format of field's content e.g type prefix, may change so parse it
28026 @c at your own risk. Yes, in general?
28028 @c The order of fields may change? Shouldn't really matter but it might
28029 @c resolve inconsistencies.
28032 If the changes are likely to break front ends, the MI version level
28033 will be increased by one. The new versions of the MI protocol are not compatible
28034 with the old versions. Old versions of MI remain available, allowing front ends
28035 to keep using them until they are modified to use the latest MI version.
28037 Since @code{--interpreter=mi} always points to the latest MI version, it is
28038 recommended that front ends request a specific version of MI when launching
28039 @value{GDBN} (e.g. @code{--interpreter=mi2}) to make sure they get an
28040 interpreter with the MI version they expect.
28042 The following table gives a summary of the the released versions of the MI
28043 interface: the version number, the version of GDB in which it first appeared
28044 and the breaking changes compared to the previous version.
28046 @multitable @columnfractions .05 .05 .9
28047 @headitem MI version @tab GDB version @tab Breaking changes
28064 The @code{-environment-pwd}, @code{-environment-directory} and
28065 @code{-environment-path} commands now returns values using the MI output
28066 syntax, rather than CLI output syntax.
28069 @code{-var-list-children}'s @code{children} result field is now a list, rather
28073 @code{-var-update}'s @code{changelist} result field is now a list, rather than
28085 The output of information about multi-location breakpoints has changed in the
28086 responses to the @code{-break-insert} and @code{-break-info} commands, as well
28087 as in the @code{=breakpoint-created} and @code{=breakpoint-modified} events.
28088 The multiple locations are now placed in a @code{locations} field, whose value
28094 If your front end cannot yet migrate to a more recent version of the
28095 MI protocol, you can nevertheless selectively enable specific features
28096 available in those recent MI versions, using the following commands:
28100 @item -fix-multi-location-breakpoint-output
28101 Use the output for multi-location breakpoints which was introduced by
28102 MI 3, even when using MI versions 2 or 1. This command has no
28103 effect when using MI version 3 or later.
28107 The best way to avoid unexpected changes in MI that might break your front
28108 end is to make your project known to @value{GDBN} developers and
28109 follow development on @email{gdb@@sourceware.org} and
28110 @email{gdb-patches@@sourceware.org}.
28111 @cindex mailing lists
28113 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
28114 @node GDB/MI Output Records
28115 @section @sc{gdb/mi} Output Records
28118 * GDB/MI Result Records::
28119 * GDB/MI Stream Records::
28120 * GDB/MI Async Records::
28121 * GDB/MI Breakpoint Information::
28122 * GDB/MI Frame Information::
28123 * GDB/MI Thread Information::
28124 * GDB/MI Ada Exception Information::
28127 @node GDB/MI Result Records
28128 @subsection @sc{gdb/mi} Result Records
28130 @cindex result records in @sc{gdb/mi}
28131 @cindex @sc{gdb/mi}, result records
28132 In addition to a number of out-of-band notifications, the response to a
28133 @sc{gdb/mi} command includes one of the following result indications:
28137 @item "^done" [ "," @var{results} ]
28138 The synchronous operation was successful, @code{@var{results}} are the return
28143 This result record is equivalent to @samp{^done}. Historically, it
28144 was output instead of @samp{^done} if the command has resumed the
28145 target. This behaviour is maintained for backward compatibility, but
28146 all frontends should treat @samp{^done} and @samp{^running}
28147 identically and rely on the @samp{*running} output record to determine
28148 which threads are resumed.
28152 @value{GDBN} has connected to a remote target.
28154 @item "^error" "," "msg=" @var{c-string} [ "," "code=" @var{c-string} ]
28156 The operation failed. The @code{msg=@var{c-string}} variable contains
28157 the corresponding error message.
28159 If present, the @code{code=@var{c-string}} variable provides an error
28160 code on which consumers can rely on to detect the corresponding
28161 error condition. At present, only one error code is defined:
28164 @item "undefined-command"
28165 Indicates that the command causing the error does not exist.
28170 @value{GDBN} has terminated.
28174 @node GDB/MI Stream Records
28175 @subsection @sc{gdb/mi} Stream Records
28177 @cindex @sc{gdb/mi}, stream records
28178 @cindex stream records in @sc{gdb/mi}
28179 @value{GDBN} internally maintains a number of output streams: the console, the
28180 target, and the log. The output intended for each of these streams is
28181 funneled through the @sc{gdb/mi} interface using @dfn{stream records}.
28183 Each stream record begins with a unique @dfn{prefix character} which
28184 identifies its stream (@pxref{GDB/MI Output Syntax, , @sc{gdb/mi} Output
28185 Syntax}). In addition to the prefix, each stream record contains a
28186 @code{@var{string-output}}. This is either raw text (with an implicit new
28187 line) or a quoted C string (which does not contain an implicit newline).
28190 @item "~" @var{string-output}
28191 The console output stream contains text that should be displayed in the
28192 CLI console window. It contains the textual responses to CLI commands.
28194 @item "@@" @var{string-output}
28195 The target output stream contains any textual output from the running
28196 target. This is only present when GDB's event loop is truly
28197 asynchronous, which is currently only the case for remote targets.
28199 @item "&" @var{string-output}
28200 The log stream contains debugging messages being produced by @value{GDBN}'s
28204 @node GDB/MI Async Records
28205 @subsection @sc{gdb/mi} Async Records
28207 @cindex async records in @sc{gdb/mi}
28208 @cindex @sc{gdb/mi}, async records
28209 @dfn{Async} records are used to notify the @sc{gdb/mi} client of
28210 additional changes that have occurred. Those changes can either be a
28211 consequence of @sc{gdb/mi} commands (e.g., a breakpoint modified) or a result of
28212 target activity (e.g., target stopped).
28214 The following is the list of possible async records:
28218 @item *running,thread-id="@var{thread}"
28219 The target is now running. The @var{thread} field can be the global
28220 thread ID of the the thread that is now running, and it can be
28221 @samp{all} if all threads are running. The frontend should assume
28222 that no interaction with a running thread is possible after this
28223 notification is produced. The frontend should not assume that this
28224 notification is output only once for any command. @value{GDBN} may
28225 emit this notification several times, either for different threads,
28226 because it cannot resume all threads together, or even for a single
28227 thread, if the thread must be stepped though some code before letting
28230 @item *stopped,reason="@var{reason}",thread-id="@var{id}",stopped-threads="@var{stopped}",core="@var{core}"
28231 The target has stopped. The @var{reason} field can have one of the
28235 @item breakpoint-hit
28236 A breakpoint was reached.
28237 @item watchpoint-trigger
28238 A watchpoint was triggered.
28239 @item read-watchpoint-trigger
28240 A read watchpoint was triggered.
28241 @item access-watchpoint-trigger
28242 An access watchpoint was triggered.
28243 @item function-finished
28244 An -exec-finish or similar CLI command was accomplished.
28245 @item location-reached
28246 An -exec-until or similar CLI command was accomplished.
28247 @item watchpoint-scope
28248 A watchpoint has gone out of scope.
28249 @item end-stepping-range
28250 An -exec-next, -exec-next-instruction, -exec-step, -exec-step-instruction or
28251 similar CLI command was accomplished.
28252 @item exited-signalled
28253 The inferior exited because of a signal.
28255 The inferior exited.
28256 @item exited-normally
28257 The inferior exited normally.
28258 @item signal-received
28259 A signal was received by the inferior.
28261 The inferior has stopped due to a library being loaded or unloaded.
28262 This can happen when @code{stop-on-solib-events} (@pxref{Files}) is
28263 set or when a @code{catch load} or @code{catch unload} catchpoint is
28264 in use (@pxref{Set Catchpoints}).
28266 The inferior has forked. This is reported when @code{catch fork}
28267 (@pxref{Set Catchpoints}) has been used.
28269 The inferior has vforked. This is reported in when @code{catch vfork}
28270 (@pxref{Set Catchpoints}) has been used.
28271 @item syscall-entry
28272 The inferior entered a system call. This is reported when @code{catch
28273 syscall} (@pxref{Set Catchpoints}) has been used.
28274 @item syscall-return
28275 The inferior returned from a system call. This is reported when
28276 @code{catch syscall} (@pxref{Set Catchpoints}) has been used.
28278 The inferior called @code{exec}. This is reported when @code{catch exec}
28279 (@pxref{Set Catchpoints}) has been used.
28282 The @var{id} field identifies the global thread ID of the thread
28283 that directly caused the stop -- for example by hitting a breakpoint.
28284 Depending on whether all-stop
28285 mode is in effect (@pxref{All-Stop Mode}), @value{GDBN} may either
28286 stop all threads, or only the thread that directly triggered the stop.
28287 If all threads are stopped, the @var{stopped} field will have the
28288 value of @code{"all"}. Otherwise, the value of the @var{stopped}
28289 field will be a list of thread identifiers. Presently, this list will
28290 always include a single thread, but frontend should be prepared to see
28291 several threads in the list. The @var{core} field reports the
28292 processor core on which the stop event has happened. This field may be absent
28293 if such information is not available.
28295 @item =thread-group-added,id="@var{id}"
28296 @itemx =thread-group-removed,id="@var{id}"
28297 A thread group was either added or removed. The @var{id} field
28298 contains the @value{GDBN} identifier of the thread group. When a thread
28299 group is added, it generally might not be associated with a running
28300 process. When a thread group is removed, its id becomes invalid and
28301 cannot be used in any way.
28303 @item =thread-group-started,id="@var{id}",pid="@var{pid}"
28304 A thread group became associated with a running program,
28305 either because the program was just started or the thread group
28306 was attached to a program. The @var{id} field contains the
28307 @value{GDBN} identifier of the thread group. The @var{pid} field
28308 contains process identifier, specific to the operating system.
28310 @item =thread-group-exited,id="@var{id}"[,exit-code="@var{code}"]
28311 A thread group is no longer associated with a running program,
28312 either because the program has exited, or because it was detached
28313 from. The @var{id} field contains the @value{GDBN} identifier of the
28314 thread group. The @var{code} field is the exit code of the inferior; it exists
28315 only when the inferior exited with some code.
28317 @item =thread-created,id="@var{id}",group-id="@var{gid}"
28318 @itemx =thread-exited,id="@var{id}",group-id="@var{gid}"
28319 A thread either was created, or has exited. The @var{id} field
28320 contains the global @value{GDBN} identifier of the thread. The @var{gid}
28321 field identifies the thread group this thread belongs to.
28323 @item =thread-selected,id="@var{id}"[,frame="@var{frame}"]
28324 Informs that the selected thread or frame were changed. This notification
28325 is not emitted as result of the @code{-thread-select} or
28326 @code{-stack-select-frame} commands, but is emitted whenever an MI command
28327 that is not documented to change the selected thread and frame actually
28328 changes them. In particular, invoking, directly or indirectly
28329 (via user-defined command), the CLI @code{thread} or @code{frame} commands,
28330 will generate this notification. Changing the thread or frame from another
28331 user interface (see @ref{Interpreters}) will also generate this notification.
28333 The @var{frame} field is only present if the newly selected thread is
28334 stopped. See @ref{GDB/MI Frame Information} for the format of its value.
28336 We suggest that in response to this notification, front ends
28337 highlight the selected thread and cause subsequent commands to apply to
28340 @item =library-loaded,...
28341 Reports that a new library file was loaded by the program. This
28342 notification has 5 fields---@var{id}, @var{target-name},
28343 @var{host-name}, @var{symbols-loaded} and @var{ranges}. The @var{id} field is an
28344 opaque identifier of the library. For remote debugging case,
28345 @var{target-name} and @var{host-name} fields give the name of the
28346 library file on the target, and on the host respectively. For native
28347 debugging, both those fields have the same value. The
28348 @var{symbols-loaded} field is emitted only for backward compatibility
28349 and should not be relied on to convey any useful information. The
28350 @var{thread-group} field, if present, specifies the id of the thread
28351 group in whose context the library was loaded. If the field is
28352 absent, it means the library was loaded in the context of all present
28353 thread groups. The @var{ranges} field specifies the ranges of addresses belonging
28356 @item =library-unloaded,...
28357 Reports that a library was unloaded by the program. This notification
28358 has 3 fields---@var{id}, @var{target-name} and @var{host-name} with
28359 the same meaning as for the @code{=library-loaded} notification.
28360 The @var{thread-group} field, if present, specifies the id of the
28361 thread group in whose context the library was unloaded. If the field is
28362 absent, it means the library was unloaded in the context of all present
28365 @item =traceframe-changed,num=@var{tfnum},tracepoint=@var{tpnum}
28366 @itemx =traceframe-changed,end
28367 Reports that the trace frame was changed and its new number is
28368 @var{tfnum}. The number of the tracepoint associated with this trace
28369 frame is @var{tpnum}.
28371 @item =tsv-created,name=@var{name},initial=@var{initial}
28372 Reports that the new trace state variable @var{name} is created with
28373 initial value @var{initial}.
28375 @item =tsv-deleted,name=@var{name}
28376 @itemx =tsv-deleted
28377 Reports that the trace state variable @var{name} is deleted or all
28378 trace state variables are deleted.
28380 @item =tsv-modified,name=@var{name},initial=@var{initial}[,current=@var{current}]
28381 Reports that the trace state variable @var{name} is modified with
28382 the initial value @var{initial}. The current value @var{current} of
28383 trace state variable is optional and is reported if the current
28384 value of trace state variable is known.
28386 @item =breakpoint-created,bkpt=@{...@}
28387 @itemx =breakpoint-modified,bkpt=@{...@}
28388 @itemx =breakpoint-deleted,id=@var{number}
28389 Reports that a breakpoint was created, modified, or deleted,
28390 respectively. Only user-visible breakpoints are reported to the MI
28393 The @var{bkpt} argument is of the same form as returned by the various
28394 breakpoint commands; @xref{GDB/MI Breakpoint Commands}. The
28395 @var{number} is the ordinal number of the breakpoint.
28397 Note that if a breakpoint is emitted in the result record of a
28398 command, then it will not also be emitted in an async record.
28400 @item =record-started,thread-group="@var{id}",method="@var{method}"[,format="@var{format}"]
28401 @itemx =record-stopped,thread-group="@var{id}"
28402 Execution log recording was either started or stopped on an
28403 inferior. The @var{id} is the @value{GDBN} identifier of the thread
28404 group corresponding to the affected inferior.
28406 The @var{method} field indicates the method used to record execution. If the
28407 method in use supports multiple recording formats, @var{format} will be present
28408 and contain the currently used format. @xref{Process Record and Replay},
28409 for existing method and format values.
28411 @item =cmd-param-changed,param=@var{param},value=@var{value}
28412 Reports that a parameter of the command @code{set @var{param}} is
28413 changed to @var{value}. In the multi-word @code{set} command,
28414 the @var{param} is the whole parameter list to @code{set} command.
28415 For example, In command @code{set check type on}, @var{param}
28416 is @code{check type} and @var{value} is @code{on}.
28418 @item =memory-changed,thread-group=@var{id},addr=@var{addr},len=@var{len}[,type="code"]
28419 Reports that bytes from @var{addr} to @var{data} + @var{len} were
28420 written in an inferior. The @var{id} is the identifier of the
28421 thread group corresponding to the affected inferior. The optional
28422 @code{type="code"} part is reported if the memory written to holds
28426 @node GDB/MI Breakpoint Information
28427 @subsection @sc{gdb/mi} Breakpoint Information
28429 When @value{GDBN} reports information about a breakpoint, a
28430 tracepoint, a watchpoint, or a catchpoint, it uses a tuple with the
28435 The breakpoint number.
28438 The type of the breakpoint. For ordinary breakpoints this will be
28439 @samp{breakpoint}, but many values are possible.
28442 If the type of the breakpoint is @samp{catchpoint}, then this
28443 indicates the exact type of catchpoint.
28446 This is the breakpoint disposition---either @samp{del}, meaning that
28447 the breakpoint will be deleted at the next stop, or @samp{keep},
28448 meaning that the breakpoint will not be deleted.
28451 This indicates whether the breakpoint is enabled, in which case the
28452 value is @samp{y}, or disabled, in which case the value is @samp{n}.
28453 Note that this is not the same as the field @code{enable}.
28456 The address of the breakpoint. This may be a hexidecimal number,
28457 giving the address; or the string @samp{<PENDING>}, for a pending
28458 breakpoint; or the string @samp{<MULTIPLE>}, for a breakpoint with
28459 multiple locations. This field will not be present if no address can
28460 be determined. For example, a watchpoint does not have an address.
28463 If known, the function in which the breakpoint appears.
28464 If not known, this field is not present.
28467 The name of the source file which contains this function, if known.
28468 If not known, this field is not present.
28471 The full file name of the source file which contains this function, if
28472 known. If not known, this field is not present.
28475 The line number at which this breakpoint appears, if known.
28476 If not known, this field is not present.
28479 If the source file is not known, this field may be provided. If
28480 provided, this holds the address of the breakpoint, possibly followed
28484 If this breakpoint is pending, this field is present and holds the
28485 text used to set the breakpoint, as entered by the user.
28488 Where this breakpoint's condition is evaluated, either @samp{host} or
28492 If this is a thread-specific breakpoint, then this identifies the
28493 thread in which the breakpoint can trigger.
28496 If this breakpoint is restricted to a particular Ada task, then this
28497 field will hold the task identifier.
28500 If the breakpoint is conditional, this is the condition expression.
28503 The ignore count of the breakpoint.
28506 The enable count of the breakpoint.
28508 @item traceframe-usage
28511 @item static-tracepoint-marker-string-id
28512 For a static tracepoint, the name of the static tracepoint marker.
28515 For a masked watchpoint, this is the mask.
28518 A tracepoint's pass count.
28520 @item original-location
28521 The location of the breakpoint as originally specified by the user.
28522 This field is optional.
28525 The number of times the breakpoint has been hit.
28528 This field is only given for tracepoints. This is either @samp{y},
28529 meaning that the tracepoint is installed, or @samp{n}, meaning that it
28533 Some extra data, the exact contents of which are type-dependent.
28536 This field is present if the breakpoint has multiple locations. It is also
28537 exceptionally present if the breakpoint is enabled and has a single, disabled
28540 The value is a list of locations. The format of a location is decribed below.
28544 A location in a multi-location breakpoint is represented as a tuple with the
28550 The location number as a dotted pair, like @samp{1.2}. The first digit is the
28551 number of the parent breakpoint. The second digit is the number of the
28552 location within that breakpoint.
28555 This indicates whether the location is enabled, in which case the
28556 value is @samp{y}, or disabled, in which case the value is @samp{n}.
28557 Note that this is not the same as the field @code{enable}.
28560 The address of this location as an hexidecimal number.
28563 If known, the function in which the location appears.
28564 If not known, this field is not present.
28567 The name of the source file which contains this location, if known.
28568 If not known, this field is not present.
28571 The full file name of the source file which contains this location, if
28572 known. If not known, this field is not present.
28575 The line number at which this location appears, if known.
28576 If not known, this field is not present.
28578 @item thread-groups
28579 The thread groups this location is in.
28583 For example, here is what the output of @code{-break-insert}
28584 (@pxref{GDB/MI Breakpoint Commands}) might be:
28587 -> -break-insert main
28588 <- ^done,bkpt=@{number="1",type="breakpoint",disp="keep",
28589 enabled="y",addr="0x08048564",func="main",file="myprog.c",
28590 fullname="/home/nickrob/myprog.c",line="68",thread-groups=["i1"],
28595 @node GDB/MI Frame Information
28596 @subsection @sc{gdb/mi} Frame Information
28598 Response from many MI commands includes an information about stack
28599 frame. This information is a tuple that may have the following
28604 The level of the stack frame. The innermost frame has the level of
28605 zero. This field is always present.
28608 The name of the function corresponding to the frame. This field may
28609 be absent if @value{GDBN} is unable to determine the function name.
28612 The code address for the frame. This field is always present.
28615 The name of the source files that correspond to the frame's code
28616 address. This field may be absent.
28619 The source line corresponding to the frames' code address. This field
28623 The name of the binary file (either executable or shared library) the
28624 corresponds to the frame's code address. This field may be absent.
28628 @node GDB/MI Thread Information
28629 @subsection @sc{gdb/mi} Thread Information
28631 Whenever @value{GDBN} has to report an information about a thread, it
28632 uses a tuple with the following fields. The fields are always present unless
28637 The global numeric id assigned to the thread by @value{GDBN}.
28640 The target-specific string identifying the thread.
28643 Additional information about the thread provided by the target.
28644 It is supposed to be human-readable and not interpreted by the
28645 frontend. This field is optional.
28648 The name of the thread. If the user specified a name using the
28649 @code{thread name} command, then this name is given. Otherwise, if
28650 @value{GDBN} can extract the thread name from the target, then that
28651 name is given. If @value{GDBN} cannot find the thread name, then this
28655 The execution state of the thread, either @samp{stopped} or @samp{running},
28656 depending on whether the thread is presently running.
28659 The stack frame currently executing in the thread. This field is only present
28660 if the thread is stopped. Its format is documented in
28661 @ref{GDB/MI Frame Information}.
28664 The value of this field is an integer number of the processor core the
28665 thread was last seen on. This field is optional.
28668 @node GDB/MI Ada Exception Information
28669 @subsection @sc{gdb/mi} Ada Exception Information
28671 Whenever a @code{*stopped} record is emitted because the program
28672 stopped after hitting an exception catchpoint (@pxref{Set Catchpoints}),
28673 @value{GDBN} provides the name of the exception that was raised via
28674 the @code{exception-name} field. Also, for exceptions that were raised
28675 with an exception message, @value{GDBN} provides that message via
28676 the @code{exception-message} field.
28678 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
28679 @node GDB/MI Simple Examples
28680 @section Simple Examples of @sc{gdb/mi} Interaction
28681 @cindex @sc{gdb/mi}, simple examples
28683 This subsection presents several simple examples of interaction using
28684 the @sc{gdb/mi} interface. In these examples, @samp{->} means that the
28685 following line is passed to @sc{gdb/mi} as input, while @samp{<-} means
28686 the output received from @sc{gdb/mi}.
28688 Note the line breaks shown in the examples are here only for
28689 readability, they don't appear in the real output.
28691 @subheading Setting a Breakpoint
28693 Setting a breakpoint generates synchronous output which contains detailed
28694 information of the breakpoint.
28697 -> -break-insert main
28698 <- ^done,bkpt=@{number="1",type="breakpoint",disp="keep",
28699 enabled="y",addr="0x08048564",func="main",file="myprog.c",
28700 fullname="/home/nickrob/myprog.c",line="68",thread-groups=["i1"],
28705 @subheading Program Execution
28707 Program execution generates asynchronous records and MI gives the
28708 reason that execution stopped.
28714 <- *stopped,reason="breakpoint-hit",disp="keep",bkptno="1",thread-id="0",
28715 frame=@{addr="0x08048564",func="main",
28716 args=[@{name="argc",value="1"@},@{name="argv",value="0xbfc4d4d4"@}],
28717 file="myprog.c",fullname="/home/nickrob/myprog.c",line="68",
28718 arch="i386:x86_64"@}
28723 <- *stopped,reason="exited-normally"
28727 @subheading Quitting @value{GDBN}
28729 Quitting @value{GDBN} just prints the result class @samp{^exit}.
28737 Please note that @samp{^exit} is printed immediately, but it might
28738 take some time for @value{GDBN} to actually exit. During that time, @value{GDBN}
28739 performs necessary cleanups, including killing programs being debugged
28740 or disconnecting from debug hardware, so the frontend should wait till
28741 @value{GDBN} exits and should only forcibly kill @value{GDBN} if it
28742 fails to exit in reasonable time.
28744 @subheading A Bad Command
28746 Here's what happens if you pass a non-existent command:
28750 <- ^error,msg="Undefined MI command: rubbish"
28755 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
28756 @node GDB/MI Command Description Format
28757 @section @sc{gdb/mi} Command Description Format
28759 The remaining sections describe blocks of commands. Each block of
28760 commands is laid out in a fashion similar to this section.
28762 @subheading Motivation
28764 The motivation for this collection of commands.
28766 @subheading Introduction
28768 A brief introduction to this collection of commands as a whole.
28770 @subheading Commands
28772 For each command in the block, the following is described:
28774 @subsubheading Synopsis
28777 -command @var{args}@dots{}
28780 @subsubheading Result
28782 @subsubheading @value{GDBN} Command
28784 The corresponding @value{GDBN} CLI command(s), if any.
28786 @subsubheading Example
28788 Example(s) formatted for readability. Some of the described commands have
28789 not been implemented yet and these are labeled N.A.@: (not available).
28792 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
28793 @node GDB/MI Breakpoint Commands
28794 @section @sc{gdb/mi} Breakpoint Commands
28796 @cindex breakpoint commands for @sc{gdb/mi}
28797 @cindex @sc{gdb/mi}, breakpoint commands
28798 This section documents @sc{gdb/mi} commands for manipulating
28801 @subheading The @code{-break-after} Command
28802 @findex -break-after
28804 @subsubheading Synopsis
28807 -break-after @var{number} @var{count}
28810 The breakpoint number @var{number} is not in effect until it has been
28811 hit @var{count} times. To see how this is reflected in the output of
28812 the @samp{-break-list} command, see the description of the
28813 @samp{-break-list} command below.
28815 @subsubheading @value{GDBN} Command
28817 The corresponding @value{GDBN} command is @samp{ignore}.
28819 @subsubheading Example
28824 ^done,bkpt=@{number="1",type="breakpoint",disp="keep",
28825 enabled="y",addr="0x000100d0",func="main",file="hello.c",
28826 fullname="/home/foo/hello.c",line="5",thread-groups=["i1"],
28834 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
28835 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
28836 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
28837 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
28838 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
28839 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
28840 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
28841 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
28842 addr="0x000100d0",func="main",file="hello.c",fullname="/home/foo/hello.c",
28843 line="5",thread-groups=["i1"],times="0",ignore="3"@}]@}
28848 @subheading The @code{-break-catch} Command
28849 @findex -break-catch
28852 @subheading The @code{-break-commands} Command
28853 @findex -break-commands
28855 @subsubheading Synopsis
28858 -break-commands @var{number} [ @var{command1} ... @var{commandN} ]
28861 Specifies the CLI commands that should be executed when breakpoint
28862 @var{number} is hit. The parameters @var{command1} to @var{commandN}
28863 are the commands. If no command is specified, any previously-set
28864 commands are cleared. @xref{Break Commands}. Typical use of this
28865 functionality is tracing a program, that is, printing of values of
28866 some variables whenever breakpoint is hit and then continuing.
28868 @subsubheading @value{GDBN} Command
28870 The corresponding @value{GDBN} command is @samp{commands}.
28872 @subsubheading Example
28877 ^done,bkpt=@{number="1",type="breakpoint",disp="keep",
28878 enabled="y",addr="0x000100d0",func="main",file="hello.c",
28879 fullname="/home/foo/hello.c",line="5",thread-groups=["i1"],
28882 -break-commands 1 "print v" "continue"
28887 @subheading The @code{-break-condition} Command
28888 @findex -break-condition
28890 @subsubheading Synopsis
28893 -break-condition @var{number} @var{expr}
28896 Breakpoint @var{number} will stop the program only if the condition in
28897 @var{expr} is true. The condition becomes part of the
28898 @samp{-break-list} output (see the description of the @samp{-break-list}
28901 @subsubheading @value{GDBN} Command
28903 The corresponding @value{GDBN} command is @samp{condition}.
28905 @subsubheading Example
28909 -break-condition 1 1
28913 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
28914 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
28915 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
28916 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
28917 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
28918 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
28919 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
28920 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
28921 addr="0x000100d0",func="main",file="hello.c",fullname="/home/foo/hello.c",
28922 line="5",cond="1",thread-groups=["i1"],times="0",ignore="3"@}]@}
28926 @subheading The @code{-break-delete} Command
28927 @findex -break-delete
28929 @subsubheading Synopsis
28932 -break-delete ( @var{breakpoint} )+
28935 Delete the breakpoint(s) whose number(s) are specified in the argument
28936 list. This is obviously reflected in the breakpoint list.
28938 @subsubheading @value{GDBN} Command
28940 The corresponding @value{GDBN} command is @samp{delete}.
28942 @subsubheading Example
28950 ^done,BreakpointTable=@{nr_rows="0",nr_cols="6",
28951 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
28952 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
28953 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
28954 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
28955 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
28956 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
28961 @subheading The @code{-break-disable} Command
28962 @findex -break-disable
28964 @subsubheading Synopsis
28967 -break-disable ( @var{breakpoint} )+
28970 Disable the named @var{breakpoint}(s). The field @samp{enabled} in the
28971 break list is now set to @samp{n} for the named @var{breakpoint}(s).
28973 @subsubheading @value{GDBN} Command
28975 The corresponding @value{GDBN} command is @samp{disable}.
28977 @subsubheading Example
28985 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
28986 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
28987 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
28988 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
28989 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
28990 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
28991 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
28992 body=[bkpt=@{number="2",type="breakpoint",disp="keep",enabled="n",
28993 addr="0x000100d0",func="main",file="hello.c",fullname="/home/foo/hello.c",
28994 line="5",thread-groups=["i1"],times="0"@}]@}
28998 @subheading The @code{-break-enable} Command
28999 @findex -break-enable
29001 @subsubheading Synopsis
29004 -break-enable ( @var{breakpoint} )+
29007 Enable (previously disabled) @var{breakpoint}(s).
29009 @subsubheading @value{GDBN} Command
29011 The corresponding @value{GDBN} command is @samp{enable}.
29013 @subsubheading Example
29021 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
29022 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
29023 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
29024 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
29025 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
29026 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
29027 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
29028 body=[bkpt=@{number="2",type="breakpoint",disp="keep",enabled="y",
29029 addr="0x000100d0",func="main",file="hello.c",fullname="/home/foo/hello.c",
29030 line="5",thread-groups=["i1"],times="0"@}]@}
29034 @subheading The @code{-break-info} Command
29035 @findex -break-info
29037 @subsubheading Synopsis
29040 -break-info @var{breakpoint}
29044 Get information about a single breakpoint.
29046 The result is a table of breakpoints. @xref{GDB/MI Breakpoint
29047 Information}, for details on the format of each breakpoint in the
29050 @subsubheading @value{GDBN} Command
29052 The corresponding @value{GDBN} command is @samp{info break @var{breakpoint}}.
29054 @subsubheading Example
29057 @subheading The @code{-break-insert} Command
29058 @findex -break-insert
29059 @anchor{-break-insert}
29061 @subsubheading Synopsis
29064 -break-insert [ -t ] [ -h ] [ -f ] [ -d ] [ -a ]
29065 [ -c @var{condition} ] [ -i @var{ignore-count} ]
29066 [ -p @var{thread-id} ] [ @var{location} ]
29070 If specified, @var{location}, can be one of:
29073 @item linespec location
29074 A linespec location. @xref{Linespec Locations}.
29076 @item explicit location
29077 An explicit location. @sc{gdb/mi} explicit locations are
29078 analogous to the CLI's explicit locations using the option names
29079 listed below. @xref{Explicit Locations}.
29082 @item --source @var{filename}
29083 The source file name of the location. This option requires the use
29084 of either @samp{--function} or @samp{--line}.
29086 @item --function @var{function}
29087 The name of a function or method.
29089 @item --label @var{label}
29090 The name of a label.
29092 @item --line @var{lineoffset}
29093 An absolute or relative line offset from the start of the location.
29096 @item address location
29097 An address location, *@var{address}. @xref{Address Locations}.
29101 The possible optional parameters of this command are:
29105 Insert a temporary breakpoint.
29107 Insert a hardware breakpoint.
29109 If @var{location} cannot be parsed (for example if it
29110 refers to unknown files or functions), create a pending
29111 breakpoint. Without this flag, @value{GDBN} will report
29112 an error, and won't create a breakpoint, if @var{location}
29115 Create a disabled breakpoint.
29117 Create a tracepoint. @xref{Tracepoints}. When this parameter
29118 is used together with @samp{-h}, a fast tracepoint is created.
29119 @item -c @var{condition}
29120 Make the breakpoint conditional on @var{condition}.
29121 @item -i @var{ignore-count}
29122 Initialize the @var{ignore-count}.
29123 @item -p @var{thread-id}
29124 Restrict the breakpoint to the thread with the specified global
29128 @subsubheading Result
29130 @xref{GDB/MI Breakpoint Information}, for details on the format of the
29131 resulting breakpoint.
29133 Note: this format is open to change.
29134 @c An out-of-band breakpoint instead of part of the result?
29136 @subsubheading @value{GDBN} Command
29138 The corresponding @value{GDBN} commands are @samp{break}, @samp{tbreak},
29139 @samp{hbreak}, and @samp{thbreak}. @c and @samp{rbreak}.
29141 @subsubheading Example
29146 ^done,bkpt=@{number="1",addr="0x0001072c",file="recursive2.c",
29147 fullname="/home/foo/recursive2.c,line="4",thread-groups=["i1"],
29150 -break-insert -t foo
29151 ^done,bkpt=@{number="2",addr="0x00010774",file="recursive2.c",
29152 fullname="/home/foo/recursive2.c,line="11",thread-groups=["i1"],
29156 ^done,BreakpointTable=@{nr_rows="2",nr_cols="6",
29157 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
29158 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
29159 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
29160 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
29161 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
29162 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
29163 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
29164 addr="0x0001072c", func="main",file="recursive2.c",
29165 fullname="/home/foo/recursive2.c,"line="4",thread-groups=["i1"],
29167 bkpt=@{number="2",type="breakpoint",disp="del",enabled="y",
29168 addr="0x00010774",func="foo",file="recursive2.c",
29169 fullname="/home/foo/recursive2.c",line="11",thread-groups=["i1"],
29172 @c -break-insert -r foo.*
29173 @c ~int foo(int, int);
29174 @c ^done,bkpt=@{number="3",addr="0x00010774",file="recursive2.c,
29175 @c "fullname="/home/foo/recursive2.c",line="11",thread-groups=["i1"],
29180 @subheading The @code{-dprintf-insert} Command
29181 @findex -dprintf-insert
29183 @subsubheading Synopsis
29186 -dprintf-insert [ -t ] [ -f ] [ -d ]
29187 [ -c @var{condition} ] [ -i @var{ignore-count} ]
29188 [ -p @var{thread-id} ] [ @var{location} ] [ @var{format} ]
29193 If supplied, @var{location} may be specified the same way as for
29194 the @code{-break-insert} command. @xref{-break-insert}.
29196 The possible optional parameters of this command are:
29200 Insert a temporary breakpoint.
29202 If @var{location} cannot be parsed (for example, if it
29203 refers to unknown files or functions), create a pending
29204 breakpoint. Without this flag, @value{GDBN} will report
29205 an error, and won't create a breakpoint, if @var{location}
29208 Create a disabled breakpoint.
29209 @item -c @var{condition}
29210 Make the breakpoint conditional on @var{condition}.
29211 @item -i @var{ignore-count}
29212 Set the ignore count of the breakpoint (@pxref{Conditions, ignore count})
29213 to @var{ignore-count}.
29214 @item -p @var{thread-id}
29215 Restrict the breakpoint to the thread with the specified global
29219 @subsubheading Result
29221 @xref{GDB/MI Breakpoint Information}, for details on the format of the
29222 resulting breakpoint.
29224 @c An out-of-band breakpoint instead of part of the result?
29226 @subsubheading @value{GDBN} Command
29228 The corresponding @value{GDBN} command is @samp{dprintf}.
29230 @subsubheading Example
29234 4-dprintf-insert foo "At foo entry\n"
29235 4^done,bkpt=@{number="1",type="dprintf",disp="keep",enabled="y",
29236 addr="0x000000000040061b",func="foo",file="mi-dprintf.c",
29237 fullname="mi-dprintf.c",line="25",thread-groups=["i1"],
29238 times="0",script=@{"printf \"At foo entry\\n\"","continue"@},
29239 original-location="foo"@}
29241 5-dprintf-insert 26 "arg=%d, g=%d\n" arg g
29242 5^done,bkpt=@{number="2",type="dprintf",disp="keep",enabled="y",
29243 addr="0x000000000040062a",func="foo",file="mi-dprintf.c",
29244 fullname="mi-dprintf.c",line="26",thread-groups=["i1"],
29245 times="0",script=@{"printf \"arg=%d, g=%d\\n\", arg, g","continue"@},
29246 original-location="mi-dprintf.c:26"@}
29250 @subheading The @code{-break-list} Command
29251 @findex -break-list
29253 @subsubheading Synopsis
29259 Displays the list of inserted breakpoints, showing the following fields:
29263 number of the breakpoint
29265 type of the breakpoint: @samp{breakpoint} or @samp{watchpoint}
29267 should the breakpoint be deleted or disabled when it is hit: @samp{keep}
29270 is the breakpoint enabled or no: @samp{y} or @samp{n}
29272 memory location at which the breakpoint is set
29274 logical location of the breakpoint, expressed by function name, file
29276 @item Thread-groups
29277 list of thread groups to which this breakpoint applies
29279 number of times the breakpoint has been hit
29282 If there are no breakpoints or watchpoints, the @code{BreakpointTable}
29283 @code{body} field is an empty list.
29285 @subsubheading @value{GDBN} Command
29287 The corresponding @value{GDBN} command is @samp{info break}.
29289 @subsubheading Example
29294 ^done,BreakpointTable=@{nr_rows="2",nr_cols="6",
29295 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
29296 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
29297 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
29298 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
29299 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
29300 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
29301 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
29302 addr="0x000100d0",func="main",file="hello.c",line="5",thread-groups=["i1"],
29304 bkpt=@{number="2",type="breakpoint",disp="keep",enabled="y",
29305 addr="0x00010114",func="foo",file="hello.c",fullname="/home/foo/hello.c",
29306 line="13",thread-groups=["i1"],times="0"@}]@}
29310 Here's an example of the result when there are no breakpoints:
29315 ^done,BreakpointTable=@{nr_rows="0",nr_cols="6",
29316 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
29317 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
29318 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
29319 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
29320 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
29321 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
29326 @subheading The @code{-break-passcount} Command
29327 @findex -break-passcount
29329 @subsubheading Synopsis
29332 -break-passcount @var{tracepoint-number} @var{passcount}
29335 Set the passcount for tracepoint @var{tracepoint-number} to
29336 @var{passcount}. If the breakpoint referred to by @var{tracepoint-number}
29337 is not a tracepoint, error is emitted. This corresponds to CLI
29338 command @samp{passcount}.
29340 @subheading The @code{-break-watch} Command
29341 @findex -break-watch
29343 @subsubheading Synopsis
29346 -break-watch [ -a | -r ]
29349 Create a watchpoint. With the @samp{-a} option it will create an
29350 @dfn{access} watchpoint, i.e., a watchpoint that triggers either on a
29351 read from or on a write to the memory location. With the @samp{-r}
29352 option, the watchpoint created is a @dfn{read} watchpoint, i.e., it will
29353 trigger only when the memory location is accessed for reading. Without
29354 either of the options, the watchpoint created is a regular watchpoint,
29355 i.e., it will trigger when the memory location is accessed for writing.
29356 @xref{Set Watchpoints, , Setting Watchpoints}.
29358 Note that @samp{-break-list} will report a single list of watchpoints and
29359 breakpoints inserted.
29361 @subsubheading @value{GDBN} Command
29363 The corresponding @value{GDBN} commands are @samp{watch}, @samp{awatch}, and
29366 @subsubheading Example
29368 Setting a watchpoint on a variable in the @code{main} function:
29373 ^done,wpt=@{number="2",exp="x"@}
29378 *stopped,reason="watchpoint-trigger",wpt=@{number="2",exp="x"@},
29379 value=@{old="-268439212",new="55"@},
29380 frame=@{func="main",args=[],file="recursive2.c",
29381 fullname="/home/foo/bar/recursive2.c",line="5",arch="i386:x86_64"@}
29385 Setting a watchpoint on a variable local to a function. @value{GDBN} will stop
29386 the program execution twice: first for the variable changing value, then
29387 for the watchpoint going out of scope.
29392 ^done,wpt=@{number="5",exp="C"@}
29397 *stopped,reason="watchpoint-trigger",
29398 wpt=@{number="5",exp="C"@},value=@{old="-276895068",new="3"@},
29399 frame=@{func="callee4",args=[],
29400 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
29401 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="13",
29402 arch="i386:x86_64"@}
29407 *stopped,reason="watchpoint-scope",wpnum="5",
29408 frame=@{func="callee3",args=[@{name="strarg",
29409 value="0x11940 \"A string argument.\""@}],
29410 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
29411 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="18",
29412 arch="i386:x86_64"@}
29416 Listing breakpoints and watchpoints, at different points in the program
29417 execution. Note that once the watchpoint goes out of scope, it is
29423 ^done,wpt=@{number="2",exp="C"@}
29426 ^done,BreakpointTable=@{nr_rows="2",nr_cols="6",
29427 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
29428 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
29429 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
29430 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
29431 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
29432 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
29433 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
29434 addr="0x00010734",func="callee4",
29435 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
29436 fullname="/home/foo/devo/gdb/testsuite/gdb.mi/basics.c"line="8",thread-groups=["i1"],
29438 bkpt=@{number="2",type="watchpoint",disp="keep",
29439 enabled="y",addr="",what="C",thread-groups=["i1"],times="0"@}]@}
29444 *stopped,reason="watchpoint-trigger",wpt=@{number="2",exp="C"@},
29445 value=@{old="-276895068",new="3"@},
29446 frame=@{func="callee4",args=[],
29447 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
29448 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="13",
29449 arch="i386:x86_64"@}
29452 ^done,BreakpointTable=@{nr_rows="2",nr_cols="6",
29453 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
29454 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
29455 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
29456 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
29457 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
29458 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
29459 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
29460 addr="0x00010734",func="callee4",
29461 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
29462 fullname="/home/foo/devo/gdb/testsuite/gdb.mi/basics.c",line="8",thread-groups=["i1"],
29464 bkpt=@{number="2",type="watchpoint",disp="keep",
29465 enabled="y",addr="",what="C",thread-groups=["i1"],times="-5"@}]@}
29469 ^done,reason="watchpoint-scope",wpnum="2",
29470 frame=@{func="callee3",args=[@{name="strarg",
29471 value="0x11940 \"A string argument.\""@}],
29472 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
29473 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="18",
29474 arch="i386:x86_64"@}
29477 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
29478 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
29479 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
29480 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
29481 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
29482 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
29483 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
29484 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
29485 addr="0x00010734",func="callee4",
29486 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
29487 fullname="/home/foo/devo/gdb/testsuite/gdb.mi/basics.c",line="8",
29488 thread-groups=["i1"],times="1"@}]@}
29493 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
29494 @node GDB/MI Catchpoint Commands
29495 @section @sc{gdb/mi} Catchpoint Commands
29497 This section documents @sc{gdb/mi} commands for manipulating
29501 * Shared Library GDB/MI Catchpoint Commands::
29502 * Ada Exception GDB/MI Catchpoint Commands::
29505 @node Shared Library GDB/MI Catchpoint Commands
29506 @subsection Shared Library @sc{gdb/mi} Catchpoints
29508 @subheading The @code{-catch-load} Command
29509 @findex -catch-load
29511 @subsubheading Synopsis
29514 -catch-load [ -t ] [ -d ] @var{regexp}
29517 Add a catchpoint for library load events. If the @samp{-t} option is used,
29518 the catchpoint is a temporary one (@pxref{Set Breaks, ,Setting
29519 Breakpoints}). If the @samp{-d} option is used, the catchpoint is created
29520 in a disabled state. The @samp{regexp} argument is a regular
29521 expression used to match the name of the loaded library.
29524 @subsubheading @value{GDBN} Command
29526 The corresponding @value{GDBN} command is @samp{catch load}.
29528 @subsubheading Example
29531 -catch-load -t foo.so
29532 ^done,bkpt=@{number="1",type="catchpoint",disp="del",enabled="y",
29533 what="load of library matching foo.so",catch-type="load",times="0"@}
29538 @subheading The @code{-catch-unload} Command
29539 @findex -catch-unload
29541 @subsubheading Synopsis
29544 -catch-unload [ -t ] [ -d ] @var{regexp}
29547 Add a catchpoint for library unload events. If the @samp{-t} option is
29548 used, the catchpoint is a temporary one (@pxref{Set Breaks, ,Setting
29549 Breakpoints}). If the @samp{-d} option is used, the catchpoint is
29550 created in a disabled state. The @samp{regexp} argument is a regular
29551 expression used to match the name of the unloaded library.
29553 @subsubheading @value{GDBN} Command
29555 The corresponding @value{GDBN} command is @samp{catch unload}.
29557 @subsubheading Example
29560 -catch-unload -d bar.so
29561 ^done,bkpt=@{number="2",type="catchpoint",disp="keep",enabled="n",
29562 what="load of library matching bar.so",catch-type="unload",times="0"@}
29566 @node Ada Exception GDB/MI Catchpoint Commands
29567 @subsection Ada Exception @sc{gdb/mi} Catchpoints
29569 The following @sc{gdb/mi} commands can be used to create catchpoints
29570 that stop the execution when Ada exceptions are being raised.
29572 @subheading The @code{-catch-assert} Command
29573 @findex -catch-assert
29575 @subsubheading Synopsis
29578 -catch-assert [ -c @var{condition}] [ -d ] [ -t ]
29581 Add a catchpoint for failed Ada assertions.
29583 The possible optional parameters for this command are:
29586 @item -c @var{condition}
29587 Make the catchpoint conditional on @var{condition}.
29589 Create a disabled catchpoint.
29591 Create a temporary catchpoint.
29594 @subsubheading @value{GDBN} Command
29596 The corresponding @value{GDBN} command is @samp{catch assert}.
29598 @subsubheading Example
29602 ^done,bkptno="5",bkpt=@{number="5",type="breakpoint",disp="keep",
29603 enabled="y",addr="0x0000000000404888",what="failed Ada assertions",
29604 thread-groups=["i1"],times="0",
29605 original-location="__gnat_debug_raise_assert_failure"@}
29609 @subheading The @code{-catch-exception} Command
29610 @findex -catch-exception
29612 @subsubheading Synopsis
29615 -catch-exception [ -c @var{condition}] [ -d ] [ -e @var{exception-name} ]
29619 Add a catchpoint stopping when Ada exceptions are raised.
29620 By default, the command stops the program when any Ada exception
29621 gets raised. But it is also possible, by using some of the
29622 optional parameters described below, to create more selective
29625 The possible optional parameters for this command are:
29628 @item -c @var{condition}
29629 Make the catchpoint conditional on @var{condition}.
29631 Create a disabled catchpoint.
29632 @item -e @var{exception-name}
29633 Only stop when @var{exception-name} is raised. This option cannot
29634 be used combined with @samp{-u}.
29636 Create a temporary catchpoint.
29638 Stop only when an unhandled exception gets raised. This option
29639 cannot be used combined with @samp{-e}.
29642 @subsubheading @value{GDBN} Command
29644 The corresponding @value{GDBN} commands are @samp{catch exception}
29645 and @samp{catch exception unhandled}.
29647 @subsubheading Example
29650 -catch-exception -e Program_Error
29651 ^done,bkptno="4",bkpt=@{number="4",type="breakpoint",disp="keep",
29652 enabled="y",addr="0x0000000000404874",
29653 what="`Program_Error' Ada exception", thread-groups=["i1"],
29654 times="0",original-location="__gnat_debug_raise_exception"@}
29658 @subheading The @code{-catch-handlers} Command
29659 @findex -catch-handlers
29661 @subsubheading Synopsis
29664 -catch-handlers [ -c @var{condition}] [ -d ] [ -e @var{exception-name} ]
29668 Add a catchpoint stopping when Ada exceptions are handled.
29669 By default, the command stops the program when any Ada exception
29670 gets handled. But it is also possible, by using some of the
29671 optional parameters described below, to create more selective
29674 The possible optional parameters for this command are:
29677 @item -c @var{condition}
29678 Make the catchpoint conditional on @var{condition}.
29680 Create a disabled catchpoint.
29681 @item -e @var{exception-name}
29682 Only stop when @var{exception-name} is handled.
29684 Create a temporary catchpoint.
29687 @subsubheading @value{GDBN} Command
29689 The corresponding @value{GDBN} command is @samp{catch handlers}.
29691 @subsubheading Example
29694 -catch-handlers -e Constraint_Error
29695 ^done,bkptno="4",bkpt=@{number="4",type="breakpoint",disp="keep",
29696 enabled="y",addr="0x0000000000402f68",
29697 what="`Constraint_Error' Ada exception handlers",thread-groups=["i1"],
29698 times="0",original-location="__gnat_begin_handler"@}
29702 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
29703 @node GDB/MI Program Context
29704 @section @sc{gdb/mi} Program Context
29706 @subheading The @code{-exec-arguments} Command
29707 @findex -exec-arguments
29710 @subsubheading Synopsis
29713 -exec-arguments @var{args}
29716 Set the inferior program arguments, to be used in the next
29719 @subsubheading @value{GDBN} Command
29721 The corresponding @value{GDBN} command is @samp{set args}.
29723 @subsubheading Example
29727 -exec-arguments -v word
29734 @subheading The @code{-exec-show-arguments} Command
29735 @findex -exec-show-arguments
29737 @subsubheading Synopsis
29740 -exec-show-arguments
29743 Print the arguments of the program.
29745 @subsubheading @value{GDBN} Command
29747 The corresponding @value{GDBN} command is @samp{show args}.
29749 @subsubheading Example
29754 @subheading The @code{-environment-cd} Command
29755 @findex -environment-cd
29757 @subsubheading Synopsis
29760 -environment-cd @var{pathdir}
29763 Set @value{GDBN}'s working directory.
29765 @subsubheading @value{GDBN} Command
29767 The corresponding @value{GDBN} command is @samp{cd}.
29769 @subsubheading Example
29773 -environment-cd /kwikemart/marge/ezannoni/flathead-dev/devo/gdb
29779 @subheading The @code{-environment-directory} Command
29780 @findex -environment-directory
29782 @subsubheading Synopsis
29785 -environment-directory [ -r ] [ @var{pathdir} ]+
29788 Add directories @var{pathdir} to beginning of search path for source files.
29789 If the @samp{-r} option is used, the search path is reset to the default
29790 search path. If directories @var{pathdir} are supplied in addition to the
29791 @samp{-r} option, the search path is first reset and then addition
29793 Multiple directories may be specified, separated by blanks. Specifying
29794 multiple directories in a single command
29795 results in the directories added to the beginning of the
29796 search path in the same order they were presented in the command.
29797 If blanks are needed as
29798 part of a directory name, double-quotes should be used around
29799 the name. In the command output, the path will show up separated
29800 by the system directory-separator character. The directory-separator
29801 character must not be used
29802 in any directory name.
29803 If no directories are specified, the current search path is displayed.
29805 @subsubheading @value{GDBN} Command
29807 The corresponding @value{GDBN} command is @samp{dir}.
29809 @subsubheading Example
29813 -environment-directory /kwikemart/marge/ezannoni/flathead-dev/devo/gdb
29814 ^done,source-path="/kwikemart/marge/ezannoni/flathead-dev/devo/gdb:$cdir:$cwd"
29816 -environment-directory ""
29817 ^done,source-path="/kwikemart/marge/ezannoni/flathead-dev/devo/gdb:$cdir:$cwd"
29819 -environment-directory -r /home/jjohnstn/src/gdb /usr/src
29820 ^done,source-path="/home/jjohnstn/src/gdb:/usr/src:$cdir:$cwd"
29822 -environment-directory -r
29823 ^done,source-path="$cdir:$cwd"
29828 @subheading The @code{-environment-path} Command
29829 @findex -environment-path
29831 @subsubheading Synopsis
29834 -environment-path [ -r ] [ @var{pathdir} ]+
29837 Add directories @var{pathdir} to beginning of search path for object files.
29838 If the @samp{-r} option is used, the search path is reset to the original
29839 search path that existed at gdb start-up. If directories @var{pathdir} are
29840 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 path is displayed.
29856 @subsubheading @value{GDBN} Command
29858 The corresponding @value{GDBN} command is @samp{path}.
29860 @subsubheading Example
29865 ^done,path="/usr/bin"
29867 -environment-path /kwikemart/marge/ezannoni/flathead-dev/ppc-eabi/gdb /bin
29868 ^done,path="/kwikemart/marge/ezannoni/flathead-dev/ppc-eabi/gdb:/bin:/usr/bin"
29870 -environment-path -r /usr/local/bin
29871 ^done,path="/usr/local/bin:/usr/bin"
29876 @subheading The @code{-environment-pwd} Command
29877 @findex -environment-pwd
29879 @subsubheading Synopsis
29885 Show the current working directory.
29887 @subsubheading @value{GDBN} Command
29889 The corresponding @value{GDBN} command is @samp{pwd}.
29891 @subsubheading Example
29896 ^done,cwd="/kwikemart/marge/ezannoni/flathead-dev/devo/gdb"
29900 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
29901 @node GDB/MI Thread Commands
29902 @section @sc{gdb/mi} Thread Commands
29905 @subheading The @code{-thread-info} Command
29906 @findex -thread-info
29908 @subsubheading Synopsis
29911 -thread-info [ @var{thread-id} ]
29914 Reports information about either a specific thread, if the
29915 @var{thread-id} parameter is present, or about all threads.
29916 @var{thread-id} is the thread's global thread ID. When printing
29917 information about all threads, also reports the global ID of the
29920 @subsubheading @value{GDBN} Command
29922 The @samp{info thread} command prints the same information
29925 @subsubheading Result
29927 The result contains the following attributes:
29931 A list of threads. The format of the elements of the list is described in
29932 @ref{GDB/MI Thread Information}.
29934 @item current-thread-id
29935 The global id of the currently selected thread. This field is omitted if there
29936 is no selected thread (for example, when the selected inferior is not running,
29937 and therefore has no threads) or if a @var{thread-id} argument was passed to
29942 @subsubheading Example
29947 @{id="2",target-id="Thread 0xb7e14b90 (LWP 21257)",
29948 frame=@{level="0",addr="0xffffe410",func="__kernel_vsyscall",
29949 args=[]@},state="running"@},
29950 @{id="1",target-id="Thread 0xb7e156b0 (LWP 21254)",
29951 frame=@{level="0",addr="0x0804891f",func="foo",
29952 args=[@{name="i",value="10"@}],
29953 file="/tmp/a.c",fullname="/tmp/a.c",line="158",arch="i386:x86_64"@},
29954 state="running"@}],
29955 current-thread-id="1"
29959 @subheading The @code{-thread-list-ids} Command
29960 @findex -thread-list-ids
29962 @subsubheading Synopsis
29968 Produces a list of the currently known global @value{GDBN} thread ids.
29969 At the end of the list it also prints the total number of such
29972 This command is retained for historical reasons, the
29973 @code{-thread-info} command should be used instead.
29975 @subsubheading @value{GDBN} Command
29977 Part of @samp{info threads} supplies the same information.
29979 @subsubheading Example
29984 ^done,thread-ids=@{thread-id="3",thread-id="2",thread-id="1"@},
29985 current-thread-id="1",number-of-threads="3"
29990 @subheading The @code{-thread-select} Command
29991 @findex -thread-select
29993 @subsubheading Synopsis
29996 -thread-select @var{thread-id}
29999 Make thread with global thread number @var{thread-id} the current
30000 thread. It prints the number of the new current thread, and the
30001 topmost frame for that thread.
30003 This command is deprecated in favor of explicitly using the
30004 @samp{--thread} option to each command.
30006 @subsubheading @value{GDBN} Command
30008 The corresponding @value{GDBN} command is @samp{thread}.
30010 @subsubheading Example
30017 *stopped,reason="end-stepping-range",thread-id="2",line="187",
30018 file="../../../devo/gdb/testsuite/gdb.threads/linux-dp.c"
30022 thread-ids=@{thread-id="3",thread-id="2",thread-id="1"@},
30023 number-of-threads="3"
30026 ^done,new-thread-id="3",
30027 frame=@{level="0",func="vprintf",
30028 args=[@{name="format",value="0x8048e9c \"%*s%c %d %c\\n\""@},
30029 @{name="arg",value="0x2"@}],file="vprintf.c",line="31",arch="i386:x86_64"@}
30033 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
30034 @node GDB/MI Ada Tasking Commands
30035 @section @sc{gdb/mi} Ada Tasking Commands
30037 @subheading The @code{-ada-task-info} Command
30038 @findex -ada-task-info
30040 @subsubheading Synopsis
30043 -ada-task-info [ @var{task-id} ]
30046 Reports information about either a specific Ada task, if the
30047 @var{task-id} parameter is present, or about all Ada tasks.
30049 @subsubheading @value{GDBN} Command
30051 The @samp{info tasks} command prints the same information
30052 about all Ada tasks (@pxref{Ada Tasks}).
30054 @subsubheading Result
30056 The result is a table of Ada tasks. The following columns are
30057 defined for each Ada task:
30061 This field exists only for the current thread. It has the value @samp{*}.
30064 The identifier that @value{GDBN} uses to refer to the Ada task.
30067 The identifier that the target uses to refer to the Ada task.
30070 The global thread identifier of the thread corresponding to the Ada
30073 This field should always exist, as Ada tasks are always implemented
30074 on top of a thread. But if @value{GDBN} cannot find this corresponding
30075 thread for any reason, the field is omitted.
30078 This field exists only when the task was created by another task.
30079 In this case, it provides the ID of the parent task.
30082 The base priority of the task.
30085 The current state of the task. For a detailed description of the
30086 possible states, see @ref{Ada Tasks}.
30089 The name of the task.
30093 @subsubheading Example
30097 ^done,tasks=@{nr_rows="3",nr_cols="8",
30098 hdr=[@{width="1",alignment="-1",col_name="current",colhdr=""@},
30099 @{width="3",alignment="1",col_name="id",colhdr="ID"@},
30100 @{width="9",alignment="1",col_name="task-id",colhdr="TID"@},
30101 @{width="4",alignment="1",col_name="thread-id",colhdr=""@},
30102 @{width="4",alignment="1",col_name="parent-id",colhdr="P-ID"@},
30103 @{width="3",alignment="1",col_name="priority",colhdr="Pri"@},
30104 @{width="22",alignment="-1",col_name="state",colhdr="State"@},
30105 @{width="1",alignment="2",col_name="name",colhdr="Name"@}],
30106 body=[@{current="*",id="1",task-id=" 644010",thread-id="1",priority="48",
30107 state="Child Termination Wait",name="main_task"@}]@}
30111 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
30112 @node GDB/MI Program Execution
30113 @section @sc{gdb/mi} Program Execution
30115 These are the asynchronous commands which generate the out-of-band
30116 record @samp{*stopped}. Currently @value{GDBN} only really executes
30117 asynchronously with remote targets and this interaction is mimicked in
30120 @subheading The @code{-exec-continue} Command
30121 @findex -exec-continue
30123 @subsubheading Synopsis
30126 -exec-continue [--reverse] [--all|--thread-group N]
30129 Resumes the execution of the inferior program, which will continue
30130 to execute until it reaches a debugger stop event. If the
30131 @samp{--reverse} option is specified, execution resumes in reverse until
30132 it reaches a stop event. Stop events may include
30135 breakpoints or watchpoints
30137 signals or exceptions
30139 the end of the process (or its beginning under @samp{--reverse})
30141 the end or beginning of a replay log if one is being used.
30143 In all-stop mode (@pxref{All-Stop
30144 Mode}), may resume only one thread, or all threads, depending on the
30145 value of the @samp{scheduler-locking} variable. If @samp{--all} is
30146 specified, all threads (in all inferiors) will be resumed. The @samp{--all} option is
30147 ignored in all-stop mode. If the @samp{--thread-group} options is
30148 specified, then all threads in that thread group are resumed.
30150 @subsubheading @value{GDBN} Command
30152 The corresponding @value{GDBN} corresponding is @samp{continue}.
30154 @subsubheading Example
30161 *stopped,reason="breakpoint-hit",disp="keep",bkptno="2",frame=@{
30162 func="foo",args=[],file="hello.c",fullname="/home/foo/bar/hello.c",
30163 line="13",arch="i386:x86_64"@}
30168 @subheading The @code{-exec-finish} Command
30169 @findex -exec-finish
30171 @subsubheading Synopsis
30174 -exec-finish [--reverse]
30177 Resumes the execution of the inferior program until the current
30178 function is exited. Displays the results returned by the function.
30179 If the @samp{--reverse} option is specified, resumes the reverse
30180 execution of the inferior program until the point where current
30181 function was called.
30183 @subsubheading @value{GDBN} Command
30185 The corresponding @value{GDBN} command is @samp{finish}.
30187 @subsubheading Example
30189 Function returning @code{void}.
30196 *stopped,reason="function-finished",frame=@{func="main",args=[],
30197 file="hello.c",fullname="/home/foo/bar/hello.c",line="7",arch="i386:x86_64"@}
30201 Function returning other than @code{void}. The name of the internal
30202 @value{GDBN} variable storing the result is printed, together with the
30209 *stopped,reason="function-finished",frame=@{addr="0x000107b0",func="foo",
30210 args=[@{name="a",value="1"],@{name="b",value="9"@}@},
30211 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
30212 arch="i386:x86_64"@},
30213 gdb-result-var="$1",return-value="0"
30218 @subheading The @code{-exec-interrupt} Command
30219 @findex -exec-interrupt
30221 @subsubheading Synopsis
30224 -exec-interrupt [--all|--thread-group N]
30227 Interrupts the background execution of the target. Note how the token
30228 associated with the stop message is the one for the execution command
30229 that has been interrupted. The token for the interrupt itself only
30230 appears in the @samp{^done} output. If the user is trying to
30231 interrupt a non-running program, an error message will be printed.
30233 Note that when asynchronous execution is enabled, this command is
30234 asynchronous just like other execution commands. That is, first the
30235 @samp{^done} response will be printed, and the target stop will be
30236 reported after that using the @samp{*stopped} notification.
30238 In non-stop mode, only the context thread is interrupted by default.
30239 All threads (in all inferiors) will be interrupted if the
30240 @samp{--all} option is specified. If the @samp{--thread-group}
30241 option is specified, all threads in that group will be interrupted.
30243 @subsubheading @value{GDBN} Command
30245 The corresponding @value{GDBN} command is @samp{interrupt}.
30247 @subsubheading Example
30258 111*stopped,signal-name="SIGINT",signal-meaning="Interrupt",
30259 frame=@{addr="0x00010140",func="foo",args=[],file="try.c",
30260 fullname="/home/foo/bar/try.c",line="13",arch="i386:x86_64"@}
30265 ^error,msg="mi_cmd_exec_interrupt: Inferior not executing."
30269 @subheading The @code{-exec-jump} Command
30272 @subsubheading Synopsis
30275 -exec-jump @var{location}
30278 Resumes execution of the inferior program at the location specified by
30279 parameter. @xref{Specify Location}, for a description of the
30280 different forms of @var{location}.
30282 @subsubheading @value{GDBN} Command
30284 The corresponding @value{GDBN} command is @samp{jump}.
30286 @subsubheading Example
30289 -exec-jump foo.c:10
30290 *running,thread-id="all"
30295 @subheading The @code{-exec-next} Command
30298 @subsubheading Synopsis
30301 -exec-next [--reverse]
30304 Resumes execution of the inferior program, stopping when the beginning
30305 of the next source line is reached.
30307 If the @samp{--reverse} option is specified, resumes reverse execution
30308 of the inferior program, stopping at the beginning of the previous
30309 source line. If you issue this command on the first line of a
30310 function, it will take you back to the caller of that function, to the
30311 source line where the function was called.
30314 @subsubheading @value{GDBN} Command
30316 The corresponding @value{GDBN} command is @samp{next}.
30318 @subsubheading Example
30324 *stopped,reason="end-stepping-range",line="8",file="hello.c"
30329 @subheading The @code{-exec-next-instruction} Command
30330 @findex -exec-next-instruction
30332 @subsubheading Synopsis
30335 -exec-next-instruction [--reverse]
30338 Executes one machine instruction. If the instruction is a function
30339 call, continues until the function returns. If the program stops at an
30340 instruction in the middle of a source line, the address will be
30343 If the @samp{--reverse} option is specified, resumes reverse execution
30344 of the inferior program, stopping at the previous instruction. If the
30345 previously executed instruction was a return from another function,
30346 it will continue to execute in reverse until the call to that function
30347 (from the current stack frame) is reached.
30349 @subsubheading @value{GDBN} Command
30351 The corresponding @value{GDBN} command is @samp{nexti}.
30353 @subsubheading Example
30357 -exec-next-instruction
30361 *stopped,reason="end-stepping-range",
30362 addr="0x000100d4",line="5",file="hello.c"
30367 @subheading The @code{-exec-return} Command
30368 @findex -exec-return
30370 @subsubheading Synopsis
30376 Makes current function return immediately. Doesn't execute the inferior.
30377 Displays the new current frame.
30379 @subsubheading @value{GDBN} Command
30381 The corresponding @value{GDBN} command is @samp{return}.
30383 @subsubheading Example
30387 200-break-insert callee4
30388 200^done,bkpt=@{number="1",addr="0x00010734",
30389 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",line="8"@}
30394 000*stopped,reason="breakpoint-hit",disp="keep",bkptno="1",
30395 frame=@{func="callee4",args=[],
30396 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
30397 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="8",
30398 arch="i386:x86_64"@}
30404 111^done,frame=@{level="0",func="callee3",
30405 args=[@{name="strarg",
30406 value="0x11940 \"A string argument.\""@}],
30407 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
30408 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="18",
30409 arch="i386:x86_64"@}
30414 @subheading The @code{-exec-run} Command
30417 @subsubheading Synopsis
30420 -exec-run [ --all | --thread-group N ] [ --start ]
30423 Starts execution of the inferior from the beginning. The inferior
30424 executes until either a breakpoint is encountered or the program
30425 exits. In the latter case the output will include an exit code, if
30426 the program has exited exceptionally.
30428 When neither the @samp{--all} nor the @samp{--thread-group} option
30429 is specified, the current inferior is started. If the
30430 @samp{--thread-group} option is specified, it should refer to a thread
30431 group of type @samp{process}, and that thread group will be started.
30432 If the @samp{--all} option is specified, then all inferiors will be started.
30434 Using the @samp{--start} option instructs the debugger to stop
30435 the execution at the start of the inferior's main subprogram,
30436 following the same behavior as the @code{start} command
30437 (@pxref{Starting}).
30439 @subsubheading @value{GDBN} Command
30441 The corresponding @value{GDBN} command is @samp{run}.
30443 @subsubheading Examples
30448 ^done,bkpt=@{number="1",addr="0x0001072c",file="recursive2.c",line="4"@}
30453 *stopped,reason="breakpoint-hit",disp="keep",bkptno="1",
30454 frame=@{func="main",args=[],file="recursive2.c",
30455 fullname="/home/foo/bar/recursive2.c",line="4",arch="i386:x86_64"@}
30460 Program exited normally:
30468 *stopped,reason="exited-normally"
30473 Program exited exceptionally:
30481 *stopped,reason="exited",exit-code="01"
30485 Another way the program can terminate is if it receives a signal such as
30486 @code{SIGINT}. In this case, @sc{gdb/mi} displays this:
30490 *stopped,reason="exited-signalled",signal-name="SIGINT",
30491 signal-meaning="Interrupt"
30495 @c @subheading -exec-signal
30498 @subheading The @code{-exec-step} Command
30501 @subsubheading Synopsis
30504 -exec-step [--reverse]
30507 Resumes execution of the inferior program, stopping when the beginning
30508 of the next source line is reached, if the next source line is not a
30509 function call. If it is, stop at the first instruction of the called
30510 function. If the @samp{--reverse} option is specified, resumes reverse
30511 execution of the inferior program, stopping at the beginning of the
30512 previously executed source line.
30514 @subsubheading @value{GDBN} Command
30516 The corresponding @value{GDBN} command is @samp{step}.
30518 @subsubheading Example
30520 Stepping into a function:
30526 *stopped,reason="end-stepping-range",
30527 frame=@{func="foo",args=[@{name="a",value="10"@},
30528 @{name="b",value="0"@}],file="recursive2.c",
30529 fullname="/home/foo/bar/recursive2.c",line="11",arch="i386:x86_64"@}
30539 *stopped,reason="end-stepping-range",line="14",file="recursive2.c"
30544 @subheading The @code{-exec-step-instruction} Command
30545 @findex -exec-step-instruction
30547 @subsubheading Synopsis
30550 -exec-step-instruction [--reverse]
30553 Resumes the inferior which executes one machine instruction. If the
30554 @samp{--reverse} option is specified, resumes reverse execution of the
30555 inferior program, stopping at the previously executed instruction.
30556 The output, once @value{GDBN} has stopped, will vary depending on
30557 whether we have stopped in the middle of a source line or not. In the
30558 former case, the address at which the program stopped will be printed
30561 @subsubheading @value{GDBN} Command
30563 The corresponding @value{GDBN} command is @samp{stepi}.
30565 @subsubheading Example
30569 -exec-step-instruction
30573 *stopped,reason="end-stepping-range",
30574 frame=@{func="foo",args=[],file="try.c",
30575 fullname="/home/foo/bar/try.c",line="10",arch="i386:x86_64"@}
30577 -exec-step-instruction
30581 *stopped,reason="end-stepping-range",
30582 frame=@{addr="0x000100f4",func="foo",args=[],file="try.c",
30583 fullname="/home/foo/bar/try.c",line="10",arch="i386:x86_64"@}
30588 @subheading The @code{-exec-until} Command
30589 @findex -exec-until
30591 @subsubheading Synopsis
30594 -exec-until [ @var{location} ]
30597 Executes the inferior until the @var{location} specified in the
30598 argument is reached. If there is no argument, the inferior executes
30599 until a source line greater than the current one is reached. The
30600 reason for stopping in this case will be @samp{location-reached}.
30602 @subsubheading @value{GDBN} Command
30604 The corresponding @value{GDBN} command is @samp{until}.
30606 @subsubheading Example
30610 -exec-until recursive2.c:6
30614 *stopped,reason="location-reached",frame=@{func="main",args=[],
30615 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="6",
30616 arch="i386:x86_64"@}
30621 @subheading -file-clear
30622 Is this going away????
30625 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
30626 @node GDB/MI Stack Manipulation
30627 @section @sc{gdb/mi} Stack Manipulation Commands
30629 @subheading The @code{-enable-frame-filters} Command
30630 @findex -enable-frame-filters
30633 -enable-frame-filters
30636 @value{GDBN} allows Python-based frame filters to affect the output of
30637 the MI commands relating to stack traces. As there is no way to
30638 implement this in a fully backward-compatible way, a front end must
30639 request that this functionality be enabled.
30641 Once enabled, this feature cannot be disabled.
30643 Note that if Python support has not been compiled into @value{GDBN},
30644 this command will still succeed (and do nothing).
30646 @subheading The @code{-stack-info-frame} Command
30647 @findex -stack-info-frame
30649 @subsubheading Synopsis
30655 Get info on the selected frame.
30657 @subsubheading @value{GDBN} Command
30659 The corresponding @value{GDBN} command is @samp{info frame} or @samp{frame}
30660 (without arguments).
30662 @subsubheading Example
30667 ^done,frame=@{level="1",addr="0x0001076c",func="callee3",
30668 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
30669 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="17",
30670 arch="i386:x86_64"@}
30674 @subheading The @code{-stack-info-depth} Command
30675 @findex -stack-info-depth
30677 @subsubheading Synopsis
30680 -stack-info-depth [ @var{max-depth} ]
30683 Return the depth of the stack. If the integer argument @var{max-depth}
30684 is specified, do not count beyond @var{max-depth} frames.
30686 @subsubheading @value{GDBN} Command
30688 There's no equivalent @value{GDBN} command.
30690 @subsubheading Example
30692 For a stack with frame levels 0 through 11:
30699 -stack-info-depth 4
30702 -stack-info-depth 12
30705 -stack-info-depth 11
30708 -stack-info-depth 13
30713 @anchor{-stack-list-arguments}
30714 @subheading The @code{-stack-list-arguments} Command
30715 @findex -stack-list-arguments
30717 @subsubheading Synopsis
30720 -stack-list-arguments [ --no-frame-filters ] [ --skip-unavailable ] @var{print-values}
30721 [ @var{low-frame} @var{high-frame} ]
30724 Display a list of the arguments for the frames between @var{low-frame}
30725 and @var{high-frame} (inclusive). If @var{low-frame} and
30726 @var{high-frame} are not provided, list the arguments for the whole
30727 call stack. If the two arguments are equal, show the single frame
30728 at the corresponding level. It is an error if @var{low-frame} is
30729 larger than the actual number of frames. On the other hand,
30730 @var{high-frame} may be larger than the actual number of frames, in
30731 which case only existing frames will be returned.
30733 If @var{print-values} is 0 or @code{--no-values}, print only the names of
30734 the variables; if it is 1 or @code{--all-values}, print also their
30735 values; and if it is 2 or @code{--simple-values}, print the name,
30736 type and value for simple data types, and the name and type for arrays,
30737 structures and unions. If the option @code{--no-frame-filters} is
30738 supplied, then Python frame filters will not be executed.
30740 If the @code{--skip-unavailable} option is specified, arguments that
30741 are not available are not listed. Partially available arguments
30742 are still displayed, however.
30744 Use of this command to obtain arguments in a single frame is
30745 deprecated in favor of the @samp{-stack-list-variables} command.
30747 @subsubheading @value{GDBN} Command
30749 @value{GDBN} does not have an equivalent command. @code{gdbtk} has a
30750 @samp{gdb_get_args} command which partially overlaps with the
30751 functionality of @samp{-stack-list-arguments}.
30753 @subsubheading Example
30760 frame=@{level="0",addr="0x00010734",func="callee4",
30761 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
30762 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="8",
30763 arch="i386:x86_64"@},
30764 frame=@{level="1",addr="0x0001076c",func="callee3",
30765 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
30766 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="17",
30767 arch="i386:x86_64"@},
30768 frame=@{level="2",addr="0x0001078c",func="callee2",
30769 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
30770 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="22",
30771 arch="i386:x86_64"@},
30772 frame=@{level="3",addr="0x000107b4",func="callee1",
30773 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
30774 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="27",
30775 arch="i386:x86_64"@},
30776 frame=@{level="4",addr="0x000107e0",func="main",
30777 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
30778 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="32",
30779 arch="i386:x86_64"@}]
30781 -stack-list-arguments 0
30784 frame=@{level="0",args=[]@},
30785 frame=@{level="1",args=[name="strarg"]@},
30786 frame=@{level="2",args=[name="intarg",name="strarg"]@},
30787 frame=@{level="3",args=[name="intarg",name="strarg",name="fltarg"]@},
30788 frame=@{level="4",args=[]@}]
30790 -stack-list-arguments 1
30793 frame=@{level="0",args=[]@},
30795 args=[@{name="strarg",value="0x11940 \"A string argument.\""@}]@},
30796 frame=@{level="2",args=[
30797 @{name="intarg",value="2"@},
30798 @{name="strarg",value="0x11940 \"A string argument.\""@}]@},
30799 @{frame=@{level="3",args=[
30800 @{name="intarg",value="2"@},
30801 @{name="strarg",value="0x11940 \"A string argument.\""@},
30802 @{name="fltarg",value="3.5"@}]@},
30803 frame=@{level="4",args=[]@}]
30805 -stack-list-arguments 0 2 2
30806 ^done,stack-args=[frame=@{level="2",args=[name="intarg",name="strarg"]@}]
30808 -stack-list-arguments 1 2 2
30809 ^done,stack-args=[frame=@{level="2",
30810 args=[@{name="intarg",value="2"@},
30811 @{name="strarg",value="0x11940 \"A string argument.\""@}]@}]
30815 @c @subheading -stack-list-exception-handlers
30818 @anchor{-stack-list-frames}
30819 @subheading The @code{-stack-list-frames} Command
30820 @findex -stack-list-frames
30822 @subsubheading Synopsis
30825 -stack-list-frames [ --no-frame-filters @var{low-frame} @var{high-frame} ]
30828 List the frames currently on the stack. For each frame it displays the
30833 The frame number, 0 being the topmost frame, i.e., the innermost function.
30835 The @code{$pc} value for that frame.
30839 File name of the source file where the function lives.
30840 @item @var{fullname}
30841 The full file name of the source file where the function lives.
30843 Line number corresponding to the @code{$pc}.
30845 The shared library where this function is defined. This is only given
30846 if the frame's function is not known.
30848 Frame's architecture.
30851 If invoked without arguments, this command prints a backtrace for the
30852 whole stack. If given two integer arguments, it shows the frames whose
30853 levels are between the two arguments (inclusive). If the two arguments
30854 are equal, it shows the single frame at the corresponding level. It is
30855 an error if @var{low-frame} is larger than the actual number of
30856 frames. On the other hand, @var{high-frame} may be larger than the
30857 actual number of frames, in which case only existing frames will be
30858 returned. If the option @code{--no-frame-filters} is supplied, then
30859 Python frame filters will not be executed.
30861 @subsubheading @value{GDBN} Command
30863 The corresponding @value{GDBN} commands are @samp{backtrace} and @samp{where}.
30865 @subsubheading Example
30867 Full stack backtrace:
30873 [frame=@{level="0",addr="0x0001076c",func="foo",
30874 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="11",
30875 arch="i386:x86_64"@},
30876 frame=@{level="1",addr="0x000107a4",func="foo",
30877 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
30878 arch="i386:x86_64"@},
30879 frame=@{level="2",addr="0x000107a4",func="foo",
30880 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
30881 arch="i386:x86_64"@},
30882 frame=@{level="3",addr="0x000107a4",func="foo",
30883 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
30884 arch="i386:x86_64"@},
30885 frame=@{level="4",addr="0x000107a4",func="foo",
30886 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
30887 arch="i386:x86_64"@},
30888 frame=@{level="5",addr="0x000107a4",func="foo",
30889 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
30890 arch="i386:x86_64"@},
30891 frame=@{level="6",addr="0x000107a4",func="foo",
30892 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
30893 arch="i386:x86_64"@},
30894 frame=@{level="7",addr="0x000107a4",func="foo",
30895 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
30896 arch="i386:x86_64"@},
30897 frame=@{level="8",addr="0x000107a4",func="foo",
30898 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
30899 arch="i386:x86_64"@},
30900 frame=@{level="9",addr="0x000107a4",func="foo",
30901 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
30902 arch="i386:x86_64"@},
30903 frame=@{level="10",addr="0x000107a4",func="foo",
30904 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
30905 arch="i386:x86_64"@},
30906 frame=@{level="11",addr="0x00010738",func="main",
30907 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="4",
30908 arch="i386:x86_64"@}]
30912 Show frames between @var{low_frame} and @var{high_frame}:
30916 -stack-list-frames 3 5
30918 [frame=@{level="3",addr="0x000107a4",func="foo",
30919 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
30920 arch="i386:x86_64"@},
30921 frame=@{level="4",addr="0x000107a4",func="foo",
30922 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
30923 arch="i386:x86_64"@},
30924 frame=@{level="5",addr="0x000107a4",func="foo",
30925 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
30926 arch="i386:x86_64"@}]
30930 Show a single frame:
30934 -stack-list-frames 3 3
30936 [frame=@{level="3",addr="0x000107a4",func="foo",
30937 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
30938 arch="i386:x86_64"@}]
30943 @subheading The @code{-stack-list-locals} Command
30944 @findex -stack-list-locals
30945 @anchor{-stack-list-locals}
30947 @subsubheading Synopsis
30950 -stack-list-locals [ --no-frame-filters ] [ --skip-unavailable ] @var{print-values}
30953 Display the local variable names for the selected frame. If
30954 @var{print-values} is 0 or @code{--no-values}, print only the names of
30955 the variables; if it is 1 or @code{--all-values}, print also their
30956 values; and if it is 2 or @code{--simple-values}, print the name,
30957 type and value for simple data types, and the name and type for arrays,
30958 structures and unions. In this last case, a frontend can immediately
30959 display the value of simple data types and create variable objects for
30960 other data types when the user wishes to explore their values in
30961 more detail. If the option @code{--no-frame-filters} is supplied, then
30962 Python frame filters will not be executed.
30964 If the @code{--skip-unavailable} option is specified, local variables
30965 that are not available are not listed. Partially available local
30966 variables are still displayed, however.
30968 This command is deprecated in favor of the
30969 @samp{-stack-list-variables} command.
30971 @subsubheading @value{GDBN} Command
30973 @samp{info locals} in @value{GDBN}, @samp{gdb_get_locals} in @code{gdbtk}.
30975 @subsubheading Example
30979 -stack-list-locals 0
30980 ^done,locals=[name="A",name="B",name="C"]
30982 -stack-list-locals --all-values
30983 ^done,locals=[@{name="A",value="1"@},@{name="B",value="2"@},
30984 @{name="C",value="@{1, 2, 3@}"@}]
30985 -stack-list-locals --simple-values
30986 ^done,locals=[@{name="A",type="int",value="1"@},
30987 @{name="B",type="int",value="2"@},@{name="C",type="int [3]"@}]
30991 @anchor{-stack-list-variables}
30992 @subheading The @code{-stack-list-variables} Command
30993 @findex -stack-list-variables
30995 @subsubheading Synopsis
30998 -stack-list-variables [ --no-frame-filters ] [ --skip-unavailable ] @var{print-values}
31001 Display the names of local variables and function arguments for the selected frame. If
31002 @var{print-values} is 0 or @code{--no-values}, print only the names of
31003 the variables; if it is 1 or @code{--all-values}, print also their
31004 values; and if it is 2 or @code{--simple-values}, print the name,
31005 type and value for simple data types, and the name and type for arrays,
31006 structures and unions. If the option @code{--no-frame-filters} is
31007 supplied, then Python frame filters will not be executed.
31009 If the @code{--skip-unavailable} option is specified, local variables
31010 and arguments that are not available are not listed. Partially
31011 available arguments and local variables are still displayed, however.
31013 @subsubheading Example
31017 -stack-list-variables --thread 1 --frame 0 --all-values
31018 ^done,variables=[@{name="x",value="11"@},@{name="s",value="@{a = 1, b = 2@}"@}]
31023 @subheading The @code{-stack-select-frame} Command
31024 @findex -stack-select-frame
31026 @subsubheading Synopsis
31029 -stack-select-frame @var{framenum}
31032 Change the selected frame. Select a different frame @var{framenum} on
31035 This command in deprecated in favor of passing the @samp{--frame}
31036 option to every command.
31038 @subsubheading @value{GDBN} Command
31040 The corresponding @value{GDBN} commands are @samp{frame}, @samp{up},
31041 @samp{down}, @samp{select-frame}, @samp{up-silent}, and @samp{down-silent}.
31043 @subsubheading Example
31047 -stack-select-frame 2
31052 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
31053 @node GDB/MI Variable Objects
31054 @section @sc{gdb/mi} Variable Objects
31058 @subheading Motivation for Variable Objects in @sc{gdb/mi}
31060 For the implementation of a variable debugger window (locals, watched
31061 expressions, etc.), we are proposing the adaptation of the existing code
31062 used by @code{Insight}.
31064 The two main reasons for that are:
31068 It has been proven in practice (it is already on its second generation).
31071 It will shorten development time (needless to say how important it is
31075 The original interface was designed to be used by Tcl code, so it was
31076 slightly changed so it could be used through @sc{gdb/mi}. This section
31077 describes the @sc{gdb/mi} operations that will be available and gives some
31078 hints about their use.
31080 @emph{Note}: In addition to the set of operations described here, we
31081 expect the @sc{gui} implementation of a variable window to require, at
31082 least, the following operations:
31085 @item @code{-gdb-show} @code{output-radix}
31086 @item @code{-stack-list-arguments}
31087 @item @code{-stack-list-locals}
31088 @item @code{-stack-select-frame}
31093 @subheading Introduction to Variable Objects
31095 @cindex variable objects in @sc{gdb/mi}
31097 Variable objects are "object-oriented" MI interface for examining and
31098 changing values of expressions. Unlike some other MI interfaces that
31099 work with expressions, variable objects are specifically designed for
31100 simple and efficient presentation in the frontend. A variable object
31101 is identified by string name. When a variable object is created, the
31102 frontend specifies the expression for that variable object. The
31103 expression can be a simple variable, or it can be an arbitrary complex
31104 expression, and can even involve CPU registers. After creating a
31105 variable object, the frontend can invoke other variable object
31106 operations---for example to obtain or change the value of a variable
31107 object, or to change display format.
31109 Variable objects have hierarchical tree structure. Any variable object
31110 that corresponds to a composite type, such as structure in C, has
31111 a number of child variable objects, for example corresponding to each
31112 element of a structure. A child variable object can itself have
31113 children, recursively. Recursion ends when we reach
31114 leaf variable objects, which always have built-in types. Child variable
31115 objects are created only by explicit request, so if a frontend
31116 is not interested in the children of a particular variable object, no
31117 child will be created.
31119 For a leaf variable object it is possible to obtain its value as a
31120 string, or set the value from a string. String value can be also
31121 obtained for a non-leaf variable object, but it's generally a string
31122 that only indicates the type of the object, and does not list its
31123 contents. Assignment to a non-leaf variable object is not allowed.
31125 A frontend does not need to read the values of all variable objects each time
31126 the program stops. Instead, MI provides an update command that lists all
31127 variable objects whose values has changed since the last update
31128 operation. This considerably reduces the amount of data that must
31129 be transferred to the frontend. As noted above, children variable
31130 objects are created on demand, and only leaf variable objects have a
31131 real value. As result, gdb will read target memory only for leaf
31132 variables that frontend has created.
31134 The automatic update is not always desirable. For example, a frontend
31135 might want to keep a value of some expression for future reference,
31136 and never update it. For another example, fetching memory is
31137 relatively slow for embedded targets, so a frontend might want
31138 to disable automatic update for the variables that are either not
31139 visible on the screen, or ``closed''. This is possible using so
31140 called ``frozen variable objects''. Such variable objects are never
31141 implicitly updated.
31143 Variable objects can be either @dfn{fixed} or @dfn{floating}. For the
31144 fixed variable object, the expression is parsed when the variable
31145 object is created, including associating identifiers to specific
31146 variables. The meaning of expression never changes. For a floating
31147 variable object the values of variables whose names appear in the
31148 expressions are re-evaluated every time in the context of the current
31149 frame. Consider this example:
31154 struct work_state state;
31161 If a fixed variable object for the @code{state} variable is created in
31162 this function, and we enter the recursive call, the variable
31163 object will report the value of @code{state} in the top-level
31164 @code{do_work} invocation. On the other hand, a floating variable
31165 object will report the value of @code{state} in the current frame.
31167 If an expression specified when creating a fixed variable object
31168 refers to a local variable, the variable object becomes bound to the
31169 thread and frame in which the variable object is created. When such
31170 variable object is updated, @value{GDBN} makes sure that the
31171 thread/frame combination the variable object is bound to still exists,
31172 and re-evaluates the variable object in context of that thread/frame.
31174 The following is the complete set of @sc{gdb/mi} operations defined to
31175 access this functionality:
31177 @multitable @columnfractions .4 .6
31178 @item @strong{Operation}
31179 @tab @strong{Description}
31181 @item @code{-enable-pretty-printing}
31182 @tab enable Python-based pretty-printing
31183 @item @code{-var-create}
31184 @tab create a variable object
31185 @item @code{-var-delete}
31186 @tab delete the variable object and/or its children
31187 @item @code{-var-set-format}
31188 @tab set the display format of this variable
31189 @item @code{-var-show-format}
31190 @tab show the display format of this variable
31191 @item @code{-var-info-num-children}
31192 @tab tells how many children this object has
31193 @item @code{-var-list-children}
31194 @tab return a list of the object's children
31195 @item @code{-var-info-type}
31196 @tab show the type of this variable object
31197 @item @code{-var-info-expression}
31198 @tab print parent-relative expression that this variable object represents
31199 @item @code{-var-info-path-expression}
31200 @tab print full expression that this variable object represents
31201 @item @code{-var-show-attributes}
31202 @tab is this variable editable? does it exist here?
31203 @item @code{-var-evaluate-expression}
31204 @tab get the value of this variable
31205 @item @code{-var-assign}
31206 @tab set the value of this variable
31207 @item @code{-var-update}
31208 @tab update the variable and its children
31209 @item @code{-var-set-frozen}
31210 @tab set frozeness attribute
31211 @item @code{-var-set-update-range}
31212 @tab set range of children to display on update
31215 In the next subsection we describe each operation in detail and suggest
31216 how it can be used.
31218 @subheading Description And Use of Operations on Variable Objects
31220 @subheading The @code{-enable-pretty-printing} Command
31221 @findex -enable-pretty-printing
31224 -enable-pretty-printing
31227 @value{GDBN} allows Python-based visualizers to affect the output of the
31228 MI variable object commands. However, because there was no way to
31229 implement this in a fully backward-compatible way, a front end must
31230 request that this functionality be enabled.
31232 Once enabled, this feature cannot be disabled.
31234 Note that if Python support has not been compiled into @value{GDBN},
31235 this command will still succeed (and do nothing).
31237 This feature is currently (as of @value{GDBN} 7.0) experimental, and
31238 may work differently in future versions of @value{GDBN}.
31240 @subheading The @code{-var-create} Command
31241 @findex -var-create
31243 @subsubheading Synopsis
31246 -var-create @{@var{name} | "-"@}
31247 @{@var{frame-addr} | "*" | "@@"@} @var{expression}
31250 This operation creates a variable object, which allows the monitoring of
31251 a variable, the result of an expression, a memory cell or a CPU
31254 The @var{name} parameter is the string by which the object can be
31255 referenced. It must be unique. If @samp{-} is specified, the varobj
31256 system will generate a string ``varNNNNNN'' automatically. It will be
31257 unique provided that one does not specify @var{name} of that format.
31258 The command fails if a duplicate name is found.
31260 The frame under which the expression should be evaluated can be
31261 specified by @var{frame-addr}. A @samp{*} indicates that the current
31262 frame should be used. A @samp{@@} indicates that a floating variable
31263 object must be created.
31265 @var{expression} is any expression valid on the current language set (must not
31266 begin with a @samp{*}), or one of the following:
31270 @samp{*@var{addr}}, where @var{addr} is the address of a memory cell
31273 @samp{*@var{addr}-@var{addr}} --- a memory address range (TBD)
31276 @samp{$@var{regname}} --- a CPU register name
31279 @cindex dynamic varobj
31280 A varobj's contents may be provided by a Python-based pretty-printer. In this
31281 case the varobj is known as a @dfn{dynamic varobj}. Dynamic varobjs
31282 have slightly different semantics in some cases. If the
31283 @code{-enable-pretty-printing} command is not sent, then @value{GDBN}
31284 will never create a dynamic varobj. This ensures backward
31285 compatibility for existing clients.
31287 @subsubheading Result
31289 This operation returns attributes of the newly-created varobj. These
31294 The name of the varobj.
31297 The number of children of the varobj. This number is not necessarily
31298 reliable for a dynamic varobj. Instead, you must examine the
31299 @samp{has_more} attribute.
31302 The varobj's scalar value. For a varobj whose type is some sort of
31303 aggregate (e.g., a @code{struct}), or for a dynamic varobj, this value
31304 will not be interesting.
31307 The varobj's type. This is a string representation of the type, as
31308 would be printed by the @value{GDBN} CLI. If @samp{print object}
31309 (@pxref{Print Settings, set print object}) is set to @code{on}, the
31310 @emph{actual} (derived) type of the object is shown rather than the
31311 @emph{declared} one.
31314 If a variable object is bound to a specific thread, then this is the
31315 thread's global identifier.
31318 For a dynamic varobj, this indicates whether there appear to be any
31319 children available. For a non-dynamic varobj, this will be 0.
31322 This attribute will be present and have the value @samp{1} if the
31323 varobj is a dynamic varobj. If the varobj is not a dynamic varobj,
31324 then this attribute will not be present.
31327 A dynamic varobj can supply a display hint to the front end. The
31328 value comes directly from the Python pretty-printer object's
31329 @code{display_hint} method. @xref{Pretty Printing API}.
31332 Typical output will look like this:
31335 name="@var{name}",numchild="@var{N}",type="@var{type}",thread-id="@var{M}",
31336 has_more="@var{has_more}"
31340 @subheading The @code{-var-delete} Command
31341 @findex -var-delete
31343 @subsubheading Synopsis
31346 -var-delete [ -c ] @var{name}
31349 Deletes a previously created variable object and all of its children.
31350 With the @samp{-c} option, just deletes the children.
31352 Returns an error if the object @var{name} is not found.
31355 @subheading The @code{-var-set-format} Command
31356 @findex -var-set-format
31358 @subsubheading Synopsis
31361 -var-set-format @var{name} @var{format-spec}
31364 Sets the output format for the value of the object @var{name} to be
31367 @anchor{-var-set-format}
31368 The syntax for the @var{format-spec} is as follows:
31371 @var{format-spec} @expansion{}
31372 @{binary | decimal | hexadecimal | octal | natural | zero-hexadecimal@}
31375 The natural format is the default format choosen automatically
31376 based on the variable type (like decimal for an @code{int}, hex
31377 for pointers, etc.).
31379 The zero-hexadecimal format has a representation similar to hexadecimal
31380 but with padding zeroes to the left of the value. For example, a 32-bit
31381 hexadecimal value of 0x1234 would be represented as 0x00001234 in the
31382 zero-hexadecimal format.
31384 For a variable with children, the format is set only on the
31385 variable itself, and the children are not affected.
31387 @subheading The @code{-var-show-format} Command
31388 @findex -var-show-format
31390 @subsubheading Synopsis
31393 -var-show-format @var{name}
31396 Returns the format used to display the value of the object @var{name}.
31399 @var{format} @expansion{}
31404 @subheading The @code{-var-info-num-children} Command
31405 @findex -var-info-num-children
31407 @subsubheading Synopsis
31410 -var-info-num-children @var{name}
31413 Returns the number of children of a variable object @var{name}:
31419 Note that this number is not completely reliable for a dynamic varobj.
31420 It will return the current number of children, but more children may
31424 @subheading The @code{-var-list-children} Command
31425 @findex -var-list-children
31427 @subsubheading Synopsis
31430 -var-list-children [@var{print-values}] @var{name} [@var{from} @var{to}]
31432 @anchor{-var-list-children}
31434 Return a list of the children of the specified variable object and
31435 create variable objects for them, if they do not already exist. With
31436 a single argument or if @var{print-values} has a value of 0 or
31437 @code{--no-values}, print only the names of the variables; if
31438 @var{print-values} is 1 or @code{--all-values}, also print their
31439 values; and if it is 2 or @code{--simple-values} print the name and
31440 value for simple data types and just the name for arrays, structures
31443 @var{from} and @var{to}, if specified, indicate the range of children
31444 to report. If @var{from} or @var{to} is less than zero, the range is
31445 reset and all children will be reported. Otherwise, children starting
31446 at @var{from} (zero-based) and up to and excluding @var{to} will be
31449 If a child range is requested, it will only affect the current call to
31450 @code{-var-list-children}, but not future calls to @code{-var-update}.
31451 For this, you must instead use @code{-var-set-update-range}. The
31452 intent of this approach is to enable a front end to implement any
31453 update approach it likes; for example, scrolling a view may cause the
31454 front end to request more children with @code{-var-list-children}, and
31455 then the front end could call @code{-var-set-update-range} with a
31456 different range to ensure that future updates are restricted to just
31459 For each child the following results are returned:
31464 Name of the variable object created for this child.
31467 The expression to be shown to the user by the front end to designate this child.
31468 For example this may be the name of a structure member.
31470 For a dynamic varobj, this value cannot be used to form an
31471 expression. There is no way to do this at all with a dynamic varobj.
31473 For C/C@t{++} structures there are several pseudo children returned to
31474 designate access qualifiers. For these pseudo children @var{exp} is
31475 @samp{public}, @samp{private}, or @samp{protected}. In this case the
31476 type and value are not present.
31478 A dynamic varobj will not report the access qualifying
31479 pseudo-children, regardless of the language. This information is not
31480 available at all with a dynamic varobj.
31483 Number of children this child has. For a dynamic varobj, this will be
31487 The type of the child. If @samp{print object}
31488 (@pxref{Print Settings, set print object}) is set to @code{on}, the
31489 @emph{actual} (derived) type of the object is shown rather than the
31490 @emph{declared} one.
31493 If values were requested, this is the value.
31496 If this variable object is associated with a thread, this is the
31497 thread's global thread id. Otherwise this result is not present.
31500 If the variable object is frozen, this variable will be present with a value of 1.
31503 A dynamic varobj can supply a display hint to the front end. The
31504 value comes directly from the Python pretty-printer object's
31505 @code{display_hint} method. @xref{Pretty Printing API}.
31508 This attribute will be present and have the value @samp{1} if the
31509 varobj is a dynamic varobj. If the varobj is not a dynamic varobj,
31510 then this attribute will not be present.
31514 The result may have its own attributes:
31518 A dynamic varobj can supply a display hint to the front end. The
31519 value comes directly from the Python pretty-printer object's
31520 @code{display_hint} method. @xref{Pretty Printing API}.
31523 This is an integer attribute which is nonzero if there are children
31524 remaining after the end of the selected range.
31527 @subsubheading Example
31531 -var-list-children n
31532 ^done,numchild=@var{n},children=[child=@{name=@var{name},exp=@var{exp},
31533 numchild=@var{n},type=@var{type}@},@r{(repeats N times)}]
31535 -var-list-children --all-values n
31536 ^done,numchild=@var{n},children=[child=@{name=@var{name},exp=@var{exp},
31537 numchild=@var{n},value=@var{value},type=@var{type}@},@r{(repeats N times)}]
31541 @subheading The @code{-var-info-type} Command
31542 @findex -var-info-type
31544 @subsubheading Synopsis
31547 -var-info-type @var{name}
31550 Returns the type of the specified variable @var{name}. The type is
31551 returned as a string in the same format as it is output by the
31555 type=@var{typename}
31559 @subheading The @code{-var-info-expression} Command
31560 @findex -var-info-expression
31562 @subsubheading Synopsis
31565 -var-info-expression @var{name}
31568 Returns a string that is suitable for presenting this
31569 variable object in user interface. The string is generally
31570 not valid expression in the current language, and cannot be evaluated.
31572 For example, if @code{a} is an array, and variable object
31573 @code{A} was created for @code{a}, then we'll get this output:
31576 (gdb) -var-info-expression A.1
31577 ^done,lang="C",exp="1"
31581 Here, the value of @code{lang} is the language name, which can be
31582 found in @ref{Supported Languages}.
31584 Note that the output of the @code{-var-list-children} command also
31585 includes those expressions, so the @code{-var-info-expression} command
31588 @subheading The @code{-var-info-path-expression} Command
31589 @findex -var-info-path-expression
31591 @subsubheading Synopsis
31594 -var-info-path-expression @var{name}
31597 Returns an expression that can be evaluated in the current
31598 context and will yield the same value that a variable object has.
31599 Compare this with the @code{-var-info-expression} command, which
31600 result can be used only for UI presentation. Typical use of
31601 the @code{-var-info-path-expression} command is creating a
31602 watchpoint from a variable object.
31604 This command is currently not valid for children of a dynamic varobj,
31605 and will give an error when invoked on one.
31607 For example, suppose @code{C} is a C@t{++} class, derived from class
31608 @code{Base}, and that the @code{Base} class has a member called
31609 @code{m_size}. Assume a variable @code{c} is has the type of
31610 @code{C} and a variable object @code{C} was created for variable
31611 @code{c}. Then, we'll get this output:
31613 (gdb) -var-info-path-expression C.Base.public.m_size
31614 ^done,path_expr=((Base)c).m_size)
31617 @subheading The @code{-var-show-attributes} Command
31618 @findex -var-show-attributes
31620 @subsubheading Synopsis
31623 -var-show-attributes @var{name}
31626 List attributes of the specified variable object @var{name}:
31629 status=@var{attr} [ ( ,@var{attr} )* ]
31633 where @var{attr} is @code{@{ @{ editable | noneditable @} | TBD @}}.
31635 @subheading The @code{-var-evaluate-expression} Command
31636 @findex -var-evaluate-expression
31638 @subsubheading Synopsis
31641 -var-evaluate-expression [-f @var{format-spec}] @var{name}
31644 Evaluates the expression that is represented by the specified variable
31645 object and returns its value as a string. The format of the string
31646 can be specified with the @samp{-f} option. The possible values of
31647 this option are the same as for @code{-var-set-format}
31648 (@pxref{-var-set-format}). If the @samp{-f} option is not specified,
31649 the current display format will be used. The current display format
31650 can be changed using the @code{-var-set-format} command.
31656 Note that one must invoke @code{-var-list-children} for a variable
31657 before the value of a child variable can be evaluated.
31659 @subheading The @code{-var-assign} Command
31660 @findex -var-assign
31662 @subsubheading Synopsis
31665 -var-assign @var{name} @var{expression}
31668 Assigns the value of @var{expression} to the variable object specified
31669 by @var{name}. The object must be @samp{editable}. If the variable's
31670 value is altered by the assign, the variable will show up in any
31671 subsequent @code{-var-update} list.
31673 @subsubheading Example
31681 ^done,changelist=[@{name="var1",in_scope="true",type_changed="false"@}]
31685 @subheading The @code{-var-update} Command
31686 @findex -var-update
31688 @subsubheading Synopsis
31691 -var-update [@var{print-values}] @{@var{name} | "*"@}
31694 Reevaluate the expressions corresponding to the variable object
31695 @var{name} and all its direct and indirect children, and return the
31696 list of variable objects whose values have changed; @var{name} must
31697 be a root variable object. Here, ``changed'' means that the result of
31698 @code{-var-evaluate-expression} before and after the
31699 @code{-var-update} is different. If @samp{*} is used as the variable
31700 object names, all existing variable objects are updated, except
31701 for frozen ones (@pxref{-var-set-frozen}). The option
31702 @var{print-values} determines whether both names and values, or just
31703 names are printed. The possible values of this option are the same
31704 as for @code{-var-list-children} (@pxref{-var-list-children}). It is
31705 recommended to use the @samp{--all-values} option, to reduce the
31706 number of MI commands needed on each program stop.
31708 With the @samp{*} parameter, if a variable object is bound to a
31709 currently running thread, it will not be updated, without any
31712 If @code{-var-set-update-range} was previously used on a varobj, then
31713 only the selected range of children will be reported.
31715 @code{-var-update} reports all the changed varobjs in a tuple named
31718 Each item in the change list is itself a tuple holding:
31722 The name of the varobj.
31725 If values were requested for this update, then this field will be
31726 present and will hold the value of the varobj.
31729 @anchor{-var-update}
31730 This field is a string which may take one of three values:
31734 The variable object's current value is valid.
31737 The variable object does not currently hold a valid value but it may
31738 hold one in the future if its associated expression comes back into
31742 The variable object no longer holds a valid value.
31743 This can occur when the executable file being debugged has changed,
31744 either through recompilation or by using the @value{GDBN} @code{file}
31745 command. The front end should normally choose to delete these variable
31749 In the future new values may be added to this list so the front should
31750 be prepared for this possibility. @xref{GDB/MI Development and Front Ends, ,@sc{GDB/MI} Development and Front Ends}.
31753 This is only present if the varobj is still valid. If the type
31754 changed, then this will be the string @samp{true}; otherwise it will
31757 When a varobj's type changes, its children are also likely to have
31758 become incorrect. Therefore, the varobj's children are automatically
31759 deleted when this attribute is @samp{true}. Also, the varobj's update
31760 range, when set using the @code{-var-set-update-range} command, is
31764 If the varobj's type changed, then this field will be present and will
31767 @item new_num_children
31768 For a dynamic varobj, if the number of children changed, or if the
31769 type changed, this will be the new number of children.
31771 The @samp{numchild} field in other varobj responses is generally not
31772 valid for a dynamic varobj -- it will show the number of children that
31773 @value{GDBN} knows about, but because dynamic varobjs lazily
31774 instantiate their children, this will not reflect the number of
31775 children which may be available.
31777 The @samp{new_num_children} attribute only reports changes to the
31778 number of children known by @value{GDBN}. This is the only way to
31779 detect whether an update has removed children (which necessarily can
31780 only happen at the end of the update range).
31783 The display hint, if any.
31786 This is an integer value, which will be 1 if there are more children
31787 available outside the varobj's update range.
31790 This attribute will be present and have the value @samp{1} if the
31791 varobj is a dynamic varobj. If the varobj is not a dynamic varobj,
31792 then this attribute will not be present.
31795 If new children were added to a dynamic varobj within the selected
31796 update range (as set by @code{-var-set-update-range}), then they will
31797 be listed in this attribute.
31800 @subsubheading Example
31807 -var-update --all-values var1
31808 ^done,changelist=[@{name="var1",value="3",in_scope="true",
31809 type_changed="false"@}]
31813 @subheading The @code{-var-set-frozen} Command
31814 @findex -var-set-frozen
31815 @anchor{-var-set-frozen}
31817 @subsubheading Synopsis
31820 -var-set-frozen @var{name} @var{flag}
31823 Set the frozenness flag on the variable object @var{name}. The
31824 @var{flag} parameter should be either @samp{1} to make the variable
31825 frozen or @samp{0} to make it unfrozen. If a variable object is
31826 frozen, then neither itself, nor any of its children, are
31827 implicitly updated by @code{-var-update} of
31828 a parent variable or by @code{-var-update *}. Only
31829 @code{-var-update} of the variable itself will update its value and
31830 values of its children. After a variable object is unfrozen, it is
31831 implicitly updated by all subsequent @code{-var-update} operations.
31832 Unfreezing a variable does not update it, only subsequent
31833 @code{-var-update} does.
31835 @subsubheading Example
31839 -var-set-frozen V 1
31844 @subheading The @code{-var-set-update-range} command
31845 @findex -var-set-update-range
31846 @anchor{-var-set-update-range}
31848 @subsubheading Synopsis
31851 -var-set-update-range @var{name} @var{from} @var{to}
31854 Set the range of children to be returned by future invocations of
31855 @code{-var-update}.
31857 @var{from} and @var{to} indicate the range of children to report. If
31858 @var{from} or @var{to} is less than zero, the range is reset and all
31859 children will be reported. Otherwise, children starting at @var{from}
31860 (zero-based) and up to and excluding @var{to} will be reported.
31862 @subsubheading Example
31866 -var-set-update-range V 1 2
31870 @subheading The @code{-var-set-visualizer} command
31871 @findex -var-set-visualizer
31872 @anchor{-var-set-visualizer}
31874 @subsubheading Synopsis
31877 -var-set-visualizer @var{name} @var{visualizer}
31880 Set a visualizer for the variable object @var{name}.
31882 @var{visualizer} is the visualizer to use. The special value
31883 @samp{None} means to disable any visualizer in use.
31885 If not @samp{None}, @var{visualizer} must be a Python expression.
31886 This expression must evaluate to a callable object which accepts a
31887 single argument. @value{GDBN} will call this object with the value of
31888 the varobj @var{name} as an argument (this is done so that the same
31889 Python pretty-printing code can be used for both the CLI and MI).
31890 When called, this object must return an object which conforms to the
31891 pretty-printing interface (@pxref{Pretty Printing API}).
31893 The pre-defined function @code{gdb.default_visualizer} may be used to
31894 select a visualizer by following the built-in process
31895 (@pxref{Selecting Pretty-Printers}). This is done automatically when
31896 a varobj is created, and so ordinarily is not needed.
31898 This feature is only available if Python support is enabled. The MI
31899 command @code{-list-features} (@pxref{GDB/MI Support Commands})
31900 can be used to check this.
31902 @subsubheading Example
31904 Resetting the visualizer:
31908 -var-set-visualizer V None
31912 Reselecting the default (type-based) visualizer:
31916 -var-set-visualizer V gdb.default_visualizer
31920 Suppose @code{SomeClass} is a visualizer class. A lambda expression
31921 can be used to instantiate this class for a varobj:
31925 -var-set-visualizer V "lambda val: SomeClass()"
31929 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
31930 @node GDB/MI Data Manipulation
31931 @section @sc{gdb/mi} Data Manipulation
31933 @cindex data manipulation, in @sc{gdb/mi}
31934 @cindex @sc{gdb/mi}, data manipulation
31935 This section describes the @sc{gdb/mi} commands that manipulate data:
31936 examine memory and registers, evaluate expressions, etc.
31938 For details about what an addressable memory unit is,
31939 @pxref{addressable memory unit}.
31941 @c REMOVED FROM THE INTERFACE.
31942 @c @subheading -data-assign
31943 @c Change the value of a program variable. Plenty of side effects.
31944 @c @subsubheading GDB Command
31946 @c @subsubheading Example
31949 @subheading The @code{-data-disassemble} Command
31950 @findex -data-disassemble
31952 @subsubheading Synopsis
31956 [ -s @var{start-addr} -e @var{end-addr} ]
31957 | [ -a @var{addr} ]
31958 | [ -f @var{filename} -l @var{linenum} [ -n @var{lines} ] ]
31966 @item @var{start-addr}
31967 is the beginning address (or @code{$pc})
31968 @item @var{end-addr}
31971 is an address anywhere within (or the name of) the function to
31972 disassemble. If an address is specified, the whole function
31973 surrounding that address will be disassembled. If a name is
31974 specified, the whole function with that name will be disassembled.
31975 @item @var{filename}
31976 is the name of the file to disassemble
31977 @item @var{linenum}
31978 is the line number to disassemble around
31980 is the number of disassembly lines to be produced. If it is -1,
31981 the whole function will be disassembled, in case no @var{end-addr} is
31982 specified. If @var{end-addr} is specified as a non-zero value, and
31983 @var{lines} is lower than the number of disassembly lines between
31984 @var{start-addr} and @var{end-addr}, only @var{lines} lines are
31985 displayed; if @var{lines} is higher than the number of lines between
31986 @var{start-addr} and @var{end-addr}, only the lines up to @var{end-addr}
31991 @item 0 disassembly only
31992 @item 1 mixed source and disassembly (deprecated)
31993 @item 2 disassembly with raw opcodes
31994 @item 3 mixed source and disassembly with raw opcodes (deprecated)
31995 @item 4 mixed source and disassembly
31996 @item 5 mixed source and disassembly with raw opcodes
31999 Modes 1 and 3 are deprecated. The output is ``source centric''
32000 which hasn't proved useful in practice.
32001 @xref{Machine Code}, for a discussion of the difference between
32002 @code{/m} and @code{/s} output of the @code{disassemble} command.
32005 @subsubheading Result
32007 The result of the @code{-data-disassemble} command will be a list named
32008 @samp{asm_insns}, the contents of this list depend on the @var{mode}
32009 used with the @code{-data-disassemble} command.
32011 For modes 0 and 2 the @samp{asm_insns} list contains tuples with the
32016 The address at which this instruction was disassembled.
32019 The name of the function this instruction is within.
32022 The decimal offset in bytes from the start of @samp{func-name}.
32025 The text disassembly for this @samp{address}.
32028 This field is only present for modes 2, 3 and 5. This contains the raw opcode
32029 bytes for the @samp{inst} field.
32033 For modes 1, 3, 4 and 5 the @samp{asm_insns} list contains tuples named
32034 @samp{src_and_asm_line}, each of which has the following fields:
32038 The line number within @samp{file}.
32041 The file name from the compilation unit. This might be an absolute
32042 file name or a relative file name depending on the compile command
32046 Absolute file name of @samp{file}. It is converted to a canonical form
32047 using the source file search path
32048 (@pxref{Source Path, ,Specifying Source Directories})
32049 and after resolving all the symbolic links.
32051 If the source file is not found this field will contain the path as
32052 present in the debug information.
32054 @item line_asm_insn
32055 This is a list of tuples containing the disassembly for @samp{line} in
32056 @samp{file}. The fields of each tuple are the same as for
32057 @code{-data-disassemble} in @var{mode} 0 and 2, so @samp{address},
32058 @samp{func-name}, @samp{offset}, @samp{inst}, and optionally
32063 Note that whatever included in the @samp{inst} field, is not
32064 manipulated directly by @sc{gdb/mi}, i.e., it is not possible to
32067 @subsubheading @value{GDBN} Command
32069 The corresponding @value{GDBN} command is @samp{disassemble}.
32071 @subsubheading Example
32073 Disassemble from the current value of @code{$pc} to @code{$pc + 20}:
32077 -data-disassemble -s $pc -e "$pc + 20" -- 0
32080 @{address="0x000107c0",func-name="main",offset="4",
32081 inst="mov 2, %o0"@},
32082 @{address="0x000107c4",func-name="main",offset="8",
32083 inst="sethi %hi(0x11800), %o2"@},
32084 @{address="0x000107c8",func-name="main",offset="12",
32085 inst="or %o2, 0x140, %o1\t! 0x11940 <_lib_version+8>"@},
32086 @{address="0x000107cc",func-name="main",offset="16",
32087 inst="sethi %hi(0x11800), %o2"@},
32088 @{address="0x000107d0",func-name="main",offset="20",
32089 inst="or %o2, 0x168, %o4\t! 0x11968 <_lib_version+48>"@}]
32093 Disassemble the whole @code{main} function. Line 32 is part of
32097 -data-disassemble -f basics.c -l 32 -- 0
32099 @{address="0x000107bc",func-name="main",offset="0",
32100 inst="save %sp, -112, %sp"@},
32101 @{address="0x000107c0",func-name="main",offset="4",
32102 inst="mov 2, %o0"@},
32103 @{address="0x000107c4",func-name="main",offset="8",
32104 inst="sethi %hi(0x11800), %o2"@},
32106 @{address="0x0001081c",func-name="main",offset="96",inst="ret "@},
32107 @{address="0x00010820",func-name="main",offset="100",inst="restore "@}]
32111 Disassemble 3 instructions from the start of @code{main}:
32115 -data-disassemble -f basics.c -l 32 -n 3 -- 0
32117 @{address="0x000107bc",func-name="main",offset="0",
32118 inst="save %sp, -112, %sp"@},
32119 @{address="0x000107c0",func-name="main",offset="4",
32120 inst="mov 2, %o0"@},
32121 @{address="0x000107c4",func-name="main",offset="8",
32122 inst="sethi %hi(0x11800), %o2"@}]
32126 Disassemble 3 instructions from the start of @code{main} in mixed mode:
32130 -data-disassemble -f basics.c -l 32 -n 3 -- 1
32132 src_and_asm_line=@{line="31",
32133 file="../../../src/gdb/testsuite/gdb.mi/basics.c",
32134 fullname="/absolute/path/to/src/gdb/testsuite/gdb.mi/basics.c",
32135 line_asm_insn=[@{address="0x000107bc",
32136 func-name="main",offset="0",inst="save %sp, -112, %sp"@}]@},
32137 src_and_asm_line=@{line="32",
32138 file="../../../src/gdb/testsuite/gdb.mi/basics.c",
32139 fullname="/absolute/path/to/src/gdb/testsuite/gdb.mi/basics.c",
32140 line_asm_insn=[@{address="0x000107c0",
32141 func-name="main",offset="4",inst="mov 2, %o0"@},
32142 @{address="0x000107c4",func-name="main",offset="8",
32143 inst="sethi %hi(0x11800), %o2"@}]@}]
32148 @subheading The @code{-data-evaluate-expression} Command
32149 @findex -data-evaluate-expression
32151 @subsubheading Synopsis
32154 -data-evaluate-expression @var{expr}
32157 Evaluate @var{expr} as an expression. The expression could contain an
32158 inferior function call. The function call will execute synchronously.
32159 If the expression contains spaces, it must be enclosed in double quotes.
32161 @subsubheading @value{GDBN} Command
32163 The corresponding @value{GDBN} commands are @samp{print}, @samp{output}, and
32164 @samp{call}. In @code{gdbtk} only, there's a corresponding
32165 @samp{gdb_eval} command.
32167 @subsubheading Example
32169 In the following example, the numbers that precede the commands are the
32170 @dfn{tokens} described in @ref{GDB/MI Command Syntax, ,@sc{gdb/mi}
32171 Command Syntax}. Notice how @sc{gdb/mi} returns the same tokens in its
32175 211-data-evaluate-expression A
32178 311-data-evaluate-expression &A
32179 311^done,value="0xefffeb7c"
32181 411-data-evaluate-expression A+3
32184 511-data-evaluate-expression "A + 3"
32190 @subheading The @code{-data-list-changed-registers} Command
32191 @findex -data-list-changed-registers
32193 @subsubheading Synopsis
32196 -data-list-changed-registers
32199 Display a list of the registers that have changed.
32201 @subsubheading @value{GDBN} Command
32203 @value{GDBN} doesn't have a direct analog for this command; @code{gdbtk}
32204 has the corresponding command @samp{gdb_changed_register_list}.
32206 @subsubheading Example
32208 On a PPC MBX board:
32216 *stopped,reason="breakpoint-hit",disp="keep",bkptno="1",frame=@{
32217 func="main",args=[],file="try.c",fullname="/home/foo/bar/try.c",
32218 line="5",arch="powerpc"@}
32220 -data-list-changed-registers
32221 ^done,changed-registers=["0","1","2","4","5","6","7","8","9",
32222 "10","11","13","14","15","16","17","18","19","20","21","22","23",
32223 "24","25","26","27","28","30","31","64","65","66","67","69"]
32228 @subheading The @code{-data-list-register-names} Command
32229 @findex -data-list-register-names
32231 @subsubheading Synopsis
32234 -data-list-register-names [ ( @var{regno} )+ ]
32237 Show a list of register names for the current target. If no arguments
32238 are given, it shows a list of the names of all the registers. If
32239 integer numbers are given as arguments, it will print a list of the
32240 names of the registers corresponding to the arguments. To ensure
32241 consistency between a register name and its number, the output list may
32242 include empty register names.
32244 @subsubheading @value{GDBN} Command
32246 @value{GDBN} does not have a command which corresponds to
32247 @samp{-data-list-register-names}. In @code{gdbtk} there is a
32248 corresponding command @samp{gdb_regnames}.
32250 @subsubheading Example
32252 For the PPC MBX board:
32255 -data-list-register-names
32256 ^done,register-names=["r0","r1","r2","r3","r4","r5","r6","r7",
32257 "r8","r9","r10","r11","r12","r13","r14","r15","r16","r17","r18",
32258 "r19","r20","r21","r22","r23","r24","r25","r26","r27","r28","r29",
32259 "r30","r31","f0","f1","f2","f3","f4","f5","f6","f7","f8","f9",
32260 "f10","f11","f12","f13","f14","f15","f16","f17","f18","f19","f20",
32261 "f21","f22","f23","f24","f25","f26","f27","f28","f29","f30","f31",
32262 "", "pc","ps","cr","lr","ctr","xer"]
32264 -data-list-register-names 1 2 3
32265 ^done,register-names=["r1","r2","r3"]
32269 @subheading The @code{-data-list-register-values} Command
32270 @findex -data-list-register-values
32272 @subsubheading Synopsis
32275 -data-list-register-values
32276 [ @code{--skip-unavailable} ] @var{fmt} [ ( @var{regno} )*]
32279 Display the registers' contents. The format according to which the
32280 registers' contents are to be returned is given by @var{fmt}, followed
32281 by an optional list of numbers specifying the registers to display. A
32282 missing list of numbers indicates that the contents of all the
32283 registers must be returned. The @code{--skip-unavailable} option
32284 indicates that only the available registers are to be returned.
32286 Allowed formats for @var{fmt} are:
32303 @subsubheading @value{GDBN} Command
32305 The corresponding @value{GDBN} commands are @samp{info reg}, @samp{info
32306 all-reg}, and (in @code{gdbtk}) @samp{gdb_fetch_registers}.
32308 @subsubheading Example
32310 For a PPC MBX board (note: line breaks are for readability only, they
32311 don't appear in the actual output):
32315 -data-list-register-values r 64 65
32316 ^done,register-values=[@{number="64",value="0xfe00a300"@},
32317 @{number="65",value="0x00029002"@}]
32319 -data-list-register-values x
32320 ^done,register-values=[@{number="0",value="0xfe0043c8"@},
32321 @{number="1",value="0x3fff88"@},@{number="2",value="0xfffffffe"@},
32322 @{number="3",value="0x0"@},@{number="4",value="0xa"@},
32323 @{number="5",value="0x3fff68"@},@{number="6",value="0x3fff58"@},
32324 @{number="7",value="0xfe011e98"@},@{number="8",value="0x2"@},
32325 @{number="9",value="0xfa202820"@},@{number="10",value="0xfa202808"@},
32326 @{number="11",value="0x1"@},@{number="12",value="0x0"@},
32327 @{number="13",value="0x4544"@},@{number="14",value="0xffdfffff"@},
32328 @{number="15",value="0xffffffff"@},@{number="16",value="0xfffffeff"@},
32329 @{number="17",value="0xefffffed"@},@{number="18",value="0xfffffffe"@},
32330 @{number="19",value="0xffffffff"@},@{number="20",value="0xffffffff"@},
32331 @{number="21",value="0xffffffff"@},@{number="22",value="0xfffffff7"@},
32332 @{number="23",value="0xffffffff"@},@{number="24",value="0xffffffff"@},
32333 @{number="25",value="0xffffffff"@},@{number="26",value="0xfffffffb"@},
32334 @{number="27",value="0xffffffff"@},@{number="28",value="0xf7bfffff"@},
32335 @{number="29",value="0x0"@},@{number="30",value="0xfe010000"@},
32336 @{number="31",value="0x0"@},@{number="32",value="0x0"@},
32337 @{number="33",value="0x0"@},@{number="34",value="0x0"@},
32338 @{number="35",value="0x0"@},@{number="36",value="0x0"@},
32339 @{number="37",value="0x0"@},@{number="38",value="0x0"@},
32340 @{number="39",value="0x0"@},@{number="40",value="0x0"@},
32341 @{number="41",value="0x0"@},@{number="42",value="0x0"@},
32342 @{number="43",value="0x0"@},@{number="44",value="0x0"@},
32343 @{number="45",value="0x0"@},@{number="46",value="0x0"@},
32344 @{number="47",value="0x0"@},@{number="48",value="0x0"@},
32345 @{number="49",value="0x0"@},@{number="50",value="0x0"@},
32346 @{number="51",value="0x0"@},@{number="52",value="0x0"@},
32347 @{number="53",value="0x0"@},@{number="54",value="0x0"@},
32348 @{number="55",value="0x0"@},@{number="56",value="0x0"@},
32349 @{number="57",value="0x0"@},@{number="58",value="0x0"@},
32350 @{number="59",value="0x0"@},@{number="60",value="0x0"@},
32351 @{number="61",value="0x0"@},@{number="62",value="0x0"@},
32352 @{number="63",value="0x0"@},@{number="64",value="0xfe00a300"@},
32353 @{number="65",value="0x29002"@},@{number="66",value="0x202f04b5"@},
32354 @{number="67",value="0xfe0043b0"@},@{number="68",value="0xfe00b3e4"@},
32355 @{number="69",value="0x20002b03"@}]
32360 @subheading The @code{-data-read-memory} Command
32361 @findex -data-read-memory
32363 This command is deprecated, use @code{-data-read-memory-bytes} instead.
32365 @subsubheading Synopsis
32368 -data-read-memory [ -o @var{byte-offset} ]
32369 @var{address} @var{word-format} @var{word-size}
32370 @var{nr-rows} @var{nr-cols} [ @var{aschar} ]
32377 @item @var{address}
32378 An expression specifying the address of the first memory word to be
32379 read. Complex expressions containing embedded white space should be
32380 quoted using the C convention.
32382 @item @var{word-format}
32383 The format to be used to print the memory words. The notation is the
32384 same as for @value{GDBN}'s @code{print} command (@pxref{Output Formats,
32387 @item @var{word-size}
32388 The size of each memory word in bytes.
32390 @item @var{nr-rows}
32391 The number of rows in the output table.
32393 @item @var{nr-cols}
32394 The number of columns in the output table.
32397 If present, indicates that each row should include an @sc{ascii} dump. The
32398 value of @var{aschar} is used as a padding character when a byte is not a
32399 member of the printable @sc{ascii} character set (printable @sc{ascii}
32400 characters are those whose code is between 32 and 126, inclusively).
32402 @item @var{byte-offset}
32403 An offset to add to the @var{address} before fetching memory.
32406 This command displays memory contents as a table of @var{nr-rows} by
32407 @var{nr-cols} words, each word being @var{word-size} bytes. In total,
32408 @code{@var{nr-rows} * @var{nr-cols} * @var{word-size}} bytes are read
32409 (returned as @samp{total-bytes}). Should less than the requested number
32410 of bytes be returned by the target, the missing words are identified
32411 using @samp{N/A}. The number of bytes read from the target is returned
32412 in @samp{nr-bytes} and the starting address used to read memory in
32415 The address of the next/previous row or page is available in
32416 @samp{next-row} and @samp{prev-row}, @samp{next-page} and
32419 @subsubheading @value{GDBN} Command
32421 The corresponding @value{GDBN} command is @samp{x}. @code{gdbtk} has
32422 @samp{gdb_get_mem} memory read command.
32424 @subsubheading Example
32426 Read six bytes of memory starting at @code{bytes+6} but then offset by
32427 @code{-6} bytes. Format as three rows of two columns. One byte per
32428 word. Display each word in hex.
32432 9-data-read-memory -o -6 -- bytes+6 x 1 3 2
32433 9^done,addr="0x00001390",nr-bytes="6",total-bytes="6",
32434 next-row="0x00001396",prev-row="0x0000138e",next-page="0x00001396",
32435 prev-page="0x0000138a",memory=[
32436 @{addr="0x00001390",data=["0x00","0x01"]@},
32437 @{addr="0x00001392",data=["0x02","0x03"]@},
32438 @{addr="0x00001394",data=["0x04","0x05"]@}]
32442 Read two bytes of memory starting at address @code{shorts + 64} and
32443 display as a single word formatted in decimal.
32447 5-data-read-memory shorts+64 d 2 1 1
32448 5^done,addr="0x00001510",nr-bytes="2",total-bytes="2",
32449 next-row="0x00001512",prev-row="0x0000150e",
32450 next-page="0x00001512",prev-page="0x0000150e",memory=[
32451 @{addr="0x00001510",data=["128"]@}]
32455 Read thirty two bytes of memory starting at @code{bytes+16} and format
32456 as eight rows of four columns. Include a string encoding with @samp{x}
32457 used as the non-printable character.
32461 4-data-read-memory bytes+16 x 1 8 4 x
32462 4^done,addr="0x000013a0",nr-bytes="32",total-bytes="32",
32463 next-row="0x000013c0",prev-row="0x0000139c",
32464 next-page="0x000013c0",prev-page="0x00001380",memory=[
32465 @{addr="0x000013a0",data=["0x10","0x11","0x12","0x13"],ascii="xxxx"@},
32466 @{addr="0x000013a4",data=["0x14","0x15","0x16","0x17"],ascii="xxxx"@},
32467 @{addr="0x000013a8",data=["0x18","0x19","0x1a","0x1b"],ascii="xxxx"@},
32468 @{addr="0x000013ac",data=["0x1c","0x1d","0x1e","0x1f"],ascii="xxxx"@},
32469 @{addr="0x000013b0",data=["0x20","0x21","0x22","0x23"],ascii=" !\"#"@},
32470 @{addr="0x000013b4",data=["0x24","0x25","0x26","0x27"],ascii="$%&'"@},
32471 @{addr="0x000013b8",data=["0x28","0x29","0x2a","0x2b"],ascii="()*+"@},
32472 @{addr="0x000013bc",data=["0x2c","0x2d","0x2e","0x2f"],ascii=",-./"@}]
32476 @subheading The @code{-data-read-memory-bytes} Command
32477 @findex -data-read-memory-bytes
32479 @subsubheading Synopsis
32482 -data-read-memory-bytes [ -o @var{offset} ]
32483 @var{address} @var{count}
32490 @item @var{address}
32491 An expression specifying the address of the first addressable memory unit
32492 to be read. Complex expressions containing embedded white space should be
32493 quoted using the C convention.
32496 The number of addressable memory units to read. This should be an integer
32500 The offset relative to @var{address} at which to start reading. This
32501 should be an integer literal. This option is provided so that a frontend
32502 is not required to first evaluate address and then perform address
32503 arithmetics itself.
32507 This command attempts to read all accessible memory regions in the
32508 specified range. First, all regions marked as unreadable in the memory
32509 map (if one is defined) will be skipped. @xref{Memory Region
32510 Attributes}. Second, @value{GDBN} will attempt to read the remaining
32511 regions. For each one, if reading full region results in an errors,
32512 @value{GDBN} will try to read a subset of the region.
32514 In general, every single memory unit in the region may be readable or not,
32515 and the only way to read every readable unit is to try a read at
32516 every address, which is not practical. Therefore, @value{GDBN} will
32517 attempt to read all accessible memory units at either beginning or the end
32518 of the region, using a binary division scheme. This heuristic works
32519 well for reading accross a memory map boundary. Note that if a region
32520 has a readable range that is neither at the beginning or the end,
32521 @value{GDBN} will not read it.
32523 The result record (@pxref{GDB/MI Result Records}) that is output of
32524 the command includes a field named @samp{memory} whose content is a
32525 list of tuples. Each tuple represent a successfully read memory block
32526 and has the following fields:
32530 The start address of the memory block, as hexadecimal literal.
32533 The end address of the memory block, as hexadecimal literal.
32536 The offset of the memory block, as hexadecimal literal, relative to
32537 the start address passed to @code{-data-read-memory-bytes}.
32540 The contents of the memory block, in hex.
32546 @subsubheading @value{GDBN} Command
32548 The corresponding @value{GDBN} command is @samp{x}.
32550 @subsubheading Example
32554 -data-read-memory-bytes &a 10
32555 ^done,memory=[@{begin="0xbffff154",offset="0x00000000",
32557 contents="01000000020000000300"@}]
32562 @subheading The @code{-data-write-memory-bytes} Command
32563 @findex -data-write-memory-bytes
32565 @subsubheading Synopsis
32568 -data-write-memory-bytes @var{address} @var{contents}
32569 -data-write-memory-bytes @var{address} @var{contents} @r{[}@var{count}@r{]}
32576 @item @var{address}
32577 An expression specifying the address of the first addressable memory unit
32578 to be written. Complex expressions containing embedded white space should
32579 be quoted using the C convention.
32581 @item @var{contents}
32582 The hex-encoded data to write. It is an error if @var{contents} does
32583 not represent an integral number of addressable memory units.
32586 Optional argument indicating the number of addressable memory units to be
32587 written. If @var{count} is greater than @var{contents}' length,
32588 @value{GDBN} will repeatedly write @var{contents} until it fills
32589 @var{count} memory units.
32593 @subsubheading @value{GDBN} Command
32595 There's no corresponding @value{GDBN} command.
32597 @subsubheading Example
32601 -data-write-memory-bytes &a "aabbccdd"
32608 -data-write-memory-bytes &a "aabbccdd" 16e
32613 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
32614 @node GDB/MI Tracepoint Commands
32615 @section @sc{gdb/mi} Tracepoint Commands
32617 The commands defined in this section implement MI support for
32618 tracepoints. For detailed introduction, see @ref{Tracepoints}.
32620 @subheading The @code{-trace-find} Command
32621 @findex -trace-find
32623 @subsubheading Synopsis
32626 -trace-find @var{mode} [@var{parameters}@dots{}]
32629 Find a trace frame using criteria defined by @var{mode} and
32630 @var{parameters}. The following table lists permissible
32631 modes and their parameters. For details of operation, see @ref{tfind}.
32636 No parameters are required. Stops examining trace frames.
32639 An integer is required as parameter. Selects tracepoint frame with
32642 @item tracepoint-number
32643 An integer is required as parameter. Finds next
32644 trace frame that corresponds to tracepoint with the specified number.
32647 An address is required as parameter. Finds
32648 next trace frame that corresponds to any tracepoint at the specified
32651 @item pc-inside-range
32652 Two addresses are required as parameters. Finds next trace
32653 frame that corresponds to a tracepoint at an address inside the
32654 specified range. Both bounds are considered to be inside the range.
32656 @item pc-outside-range
32657 Two addresses are required as parameters. Finds
32658 next trace frame that corresponds to a tracepoint at an address outside
32659 the specified range. Both bounds are considered to be inside the range.
32662 Line specification is required as parameter. @xref{Specify Location}.
32663 Finds next trace frame that corresponds to a tracepoint at
32664 the specified location.
32668 If @samp{none} was passed as @var{mode}, the response does not
32669 have fields. Otherwise, the response may have the following fields:
32673 This field has either @samp{0} or @samp{1} as the value, depending
32674 on whether a matching tracepoint was found.
32677 The index of the found traceframe. This field is present iff
32678 the @samp{found} field has value of @samp{1}.
32681 The index of the found tracepoint. This field is present iff
32682 the @samp{found} field has value of @samp{1}.
32685 The information about the frame corresponding to the found trace
32686 frame. This field is present only if a trace frame was found.
32687 @xref{GDB/MI Frame Information}, for description of this field.
32691 @subsubheading @value{GDBN} Command
32693 The corresponding @value{GDBN} command is @samp{tfind}.
32695 @subheading -trace-define-variable
32696 @findex -trace-define-variable
32698 @subsubheading Synopsis
32701 -trace-define-variable @var{name} [ @var{value} ]
32704 Create trace variable @var{name} if it does not exist. If
32705 @var{value} is specified, sets the initial value of the specified
32706 trace variable to that value. Note that the @var{name} should start
32707 with the @samp{$} character.
32709 @subsubheading @value{GDBN} Command
32711 The corresponding @value{GDBN} command is @samp{tvariable}.
32713 @subheading The @code{-trace-frame-collected} Command
32714 @findex -trace-frame-collected
32716 @subsubheading Synopsis
32719 -trace-frame-collected
32720 [--var-print-values @var{var_pval}]
32721 [--comp-print-values @var{comp_pval}]
32722 [--registers-format @var{regformat}]
32723 [--memory-contents]
32726 This command returns the set of collected objects, register names,
32727 trace state variable names, memory ranges and computed expressions
32728 that have been collected at a particular trace frame. The optional
32729 parameters to the command affect the output format in different ways.
32730 See the output description table below for more details.
32732 The reported names can be used in the normal manner to create
32733 varobjs and inspect the objects themselves. The items returned by
32734 this command are categorized so that it is clear which is a variable,
32735 which is a register, which is a trace state variable, which is a
32736 memory range and which is a computed expression.
32738 For instance, if the actions were
32740 collect myVar, myArray[myIndex], myObj.field, myPtr->field, myCount + 2
32741 collect *(int*)0xaf02bef0@@40
32745 the object collected in its entirety would be @code{myVar}. The
32746 object @code{myArray} would be partially collected, because only the
32747 element at index @code{myIndex} would be collected. The remaining
32748 objects would be computed expressions.
32750 An example output would be:
32754 -trace-frame-collected
32756 explicit-variables=[@{name="myVar",value="1"@}],
32757 computed-expressions=[@{name="myArray[myIndex]",value="0"@},
32758 @{name="myObj.field",value="0"@},
32759 @{name="myPtr->field",value="1"@},
32760 @{name="myCount + 2",value="3"@},
32761 @{name="$tvar1 + 1",value="43970027"@}],
32762 registers=[@{number="0",value="0x7fe2c6e79ec8"@},
32763 @{number="1",value="0x0"@},
32764 @{number="2",value="0x4"@},
32766 @{number="125",value="0x0"@}],
32767 tvars=[@{name="$tvar1",current="43970026"@}],
32768 memory=[@{address="0x0000000000602264",length="4"@},
32769 @{address="0x0000000000615bc0",length="4"@}]
32776 @item explicit-variables
32777 The set of objects that have been collected in their entirety (as
32778 opposed to collecting just a few elements of an array or a few struct
32779 members). For each object, its name and value are printed.
32780 The @code{--var-print-values} option affects how or whether the value
32781 field is output. If @var{var_pval} is 0, then print only the names;
32782 if it is 1, print also their values; and if it is 2, print the name,
32783 type and value for simple data types, and the name and type for
32784 arrays, structures and unions.
32786 @item computed-expressions
32787 The set of computed expressions that have been collected at the
32788 current trace frame. The @code{--comp-print-values} option affects
32789 this set like the @code{--var-print-values} option affects the
32790 @code{explicit-variables} set. See above.
32793 The registers that have been collected at the current trace frame.
32794 For each register collected, the name and current value are returned.
32795 The value is formatted according to the @code{--registers-format}
32796 option. See the @command{-data-list-register-values} command for a
32797 list of the allowed formats. The default is @samp{x}.
32800 The trace state variables that have been collected at the current
32801 trace frame. For each trace state variable collected, the name and
32802 current value are returned.
32805 The set of memory ranges that have been collected at the current trace
32806 frame. Its content is a list of tuples. Each tuple represents a
32807 collected memory range and has the following fields:
32811 The start address of the memory range, as hexadecimal literal.
32814 The length of the memory range, as decimal literal.
32817 The contents of the memory block, in hex. This field is only present
32818 if the @code{--memory-contents} option is specified.
32824 @subsubheading @value{GDBN} Command
32826 There is no corresponding @value{GDBN} command.
32828 @subsubheading Example
32830 @subheading -trace-list-variables
32831 @findex -trace-list-variables
32833 @subsubheading Synopsis
32836 -trace-list-variables
32839 Return a table of all defined trace variables. Each element of the
32840 table has the following fields:
32844 The name of the trace variable. This field is always present.
32847 The initial value. This is a 64-bit signed integer. This
32848 field is always present.
32851 The value the trace variable has at the moment. This is a 64-bit
32852 signed integer. This field is absent iff current value is
32853 not defined, for example if the trace was never run, or is
32858 @subsubheading @value{GDBN} Command
32860 The corresponding @value{GDBN} command is @samp{tvariables}.
32862 @subsubheading Example
32866 -trace-list-variables
32867 ^done,trace-variables=@{nr_rows="1",nr_cols="3",
32868 hdr=[@{width="15",alignment="-1",col_name="name",colhdr="Name"@},
32869 @{width="11",alignment="-1",col_name="initial",colhdr="Initial"@},
32870 @{width="11",alignment="-1",col_name="current",colhdr="Current"@}],
32871 body=[variable=@{name="$trace_timestamp",initial="0"@}
32872 variable=@{name="$foo",initial="10",current="15"@}]@}
32876 @subheading -trace-save
32877 @findex -trace-save
32879 @subsubheading Synopsis
32882 -trace-save [ -r ] [ -ctf ] @var{filename}
32885 Saves the collected trace data to @var{filename}. Without the
32886 @samp{-r} option, the data is downloaded from the target and saved
32887 in a local file. With the @samp{-r} option the target is asked
32888 to perform the save.
32890 By default, this command will save the trace in the tfile format. You can
32891 supply the optional @samp{-ctf} argument to save it the CTF format. See
32892 @ref{Trace Files} for more information about CTF.
32894 @subsubheading @value{GDBN} Command
32896 The corresponding @value{GDBN} command is @samp{tsave}.
32899 @subheading -trace-start
32900 @findex -trace-start
32902 @subsubheading Synopsis
32908 Starts a tracing experiment. The result of this command does not
32911 @subsubheading @value{GDBN} Command
32913 The corresponding @value{GDBN} command is @samp{tstart}.
32915 @subheading -trace-status
32916 @findex -trace-status
32918 @subsubheading Synopsis
32924 Obtains the status of a tracing experiment. The result may include
32925 the following fields:
32930 May have a value of either @samp{0}, when no tracing operations are
32931 supported, @samp{1}, when all tracing operations are supported, or
32932 @samp{file} when examining trace file. In the latter case, examining
32933 of trace frame is possible but new tracing experiement cannot be
32934 started. This field is always present.
32937 May have a value of either @samp{0} or @samp{1} depending on whether
32938 tracing experiement is in progress on target. This field is present
32939 if @samp{supported} field is not @samp{0}.
32942 Report the reason why the tracing was stopped last time. This field
32943 may be absent iff tracing was never stopped on target yet. The
32944 value of @samp{request} means the tracing was stopped as result of
32945 the @code{-trace-stop} command. The value of @samp{overflow} means
32946 the tracing buffer is full. The value of @samp{disconnection} means
32947 tracing was automatically stopped when @value{GDBN} has disconnected.
32948 The value of @samp{passcount} means tracing was stopped when a
32949 tracepoint was passed a maximal number of times for that tracepoint.
32950 This field is present if @samp{supported} field is not @samp{0}.
32952 @item stopping-tracepoint
32953 The number of tracepoint whose passcount as exceeded. This field is
32954 present iff the @samp{stop-reason} field has the value of
32958 @itemx frames-created
32959 The @samp{frames} field is a count of the total number of trace frames
32960 in the trace buffer, while @samp{frames-created} is the total created
32961 during the run, including ones that were discarded, such as when a
32962 circular trace buffer filled up. Both fields are optional.
32966 These fields tell the current size of the tracing buffer and the
32967 remaining space. These fields are optional.
32970 The value of the circular trace buffer flag. @code{1} means that the
32971 trace buffer is circular and old trace frames will be discarded if
32972 necessary to make room, @code{0} means that the trace buffer is linear
32976 The value of the disconnected tracing flag. @code{1} means that
32977 tracing will continue after @value{GDBN} disconnects, @code{0} means
32978 that the trace run will stop.
32981 The filename of the trace file being examined. This field is
32982 optional, and only present when examining a trace file.
32986 @subsubheading @value{GDBN} Command
32988 The corresponding @value{GDBN} command is @samp{tstatus}.
32990 @subheading -trace-stop
32991 @findex -trace-stop
32993 @subsubheading Synopsis
32999 Stops a tracing experiment. The result of this command has the same
33000 fields as @code{-trace-status}, except that the @samp{supported} and
33001 @samp{running} fields are not output.
33003 @subsubheading @value{GDBN} Command
33005 The corresponding @value{GDBN} command is @samp{tstop}.
33008 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
33009 @node GDB/MI Symbol Query
33010 @section @sc{gdb/mi} Symbol Query Commands
33014 @subheading The @code{-symbol-info-address} Command
33015 @findex -symbol-info-address
33017 @subsubheading Synopsis
33020 -symbol-info-address @var{symbol}
33023 Describe where @var{symbol} is stored.
33025 @subsubheading @value{GDBN} Command
33027 The corresponding @value{GDBN} command is @samp{info address}.
33029 @subsubheading Example
33033 @subheading The @code{-symbol-info-file} Command
33034 @findex -symbol-info-file
33036 @subsubheading Synopsis
33042 Show the file for the symbol.
33044 @subsubheading @value{GDBN} Command
33046 There's no equivalent @value{GDBN} command. @code{gdbtk} has
33047 @samp{gdb_find_file}.
33049 @subsubheading Example
33053 @subheading The @code{-symbol-info-function} Command
33054 @findex -symbol-info-function
33056 @subsubheading Synopsis
33059 -symbol-info-function
33062 Show which function the symbol lives in.
33064 @subsubheading @value{GDBN} Command
33066 @samp{gdb_get_function} in @code{gdbtk}.
33068 @subsubheading Example
33072 @subheading The @code{-symbol-info-line} Command
33073 @findex -symbol-info-line
33075 @subsubheading Synopsis
33081 Show the core addresses of the code for a source line.
33083 @subsubheading @value{GDBN} Command
33085 The corresponding @value{GDBN} command is @samp{info line}.
33086 @code{gdbtk} has the @samp{gdb_get_line} and @samp{gdb_get_file} commands.
33088 @subsubheading Example
33092 @subheading The @code{-symbol-info-symbol} Command
33093 @findex -symbol-info-symbol
33095 @subsubheading Synopsis
33098 -symbol-info-symbol @var{addr}
33101 Describe what symbol is at location @var{addr}.
33103 @subsubheading @value{GDBN} Command
33105 The corresponding @value{GDBN} command is @samp{info symbol}.
33107 @subsubheading Example
33111 @subheading The @code{-symbol-list-functions} Command
33112 @findex -symbol-list-functions
33114 @subsubheading Synopsis
33117 -symbol-list-functions
33120 List the functions in the executable.
33122 @subsubheading @value{GDBN} Command
33124 @samp{info functions} in @value{GDBN}, @samp{gdb_listfunc} and
33125 @samp{gdb_search} in @code{gdbtk}.
33127 @subsubheading Example
33132 @subheading The @code{-symbol-list-lines} Command
33133 @findex -symbol-list-lines
33135 @subsubheading Synopsis
33138 -symbol-list-lines @var{filename}
33141 Print the list of lines that contain code and their associated program
33142 addresses for the given source filename. The entries are sorted in
33143 ascending PC order.
33145 @subsubheading @value{GDBN} Command
33147 There is no corresponding @value{GDBN} command.
33149 @subsubheading Example
33152 -symbol-list-lines basics.c
33153 ^done,lines=[@{pc="0x08048554",line="7"@},@{pc="0x0804855a",line="8"@}]
33159 @subheading The @code{-symbol-list-types} Command
33160 @findex -symbol-list-types
33162 @subsubheading Synopsis
33168 List all the type names.
33170 @subsubheading @value{GDBN} Command
33172 The corresponding commands are @samp{info types} in @value{GDBN},
33173 @samp{gdb_search} in @code{gdbtk}.
33175 @subsubheading Example
33179 @subheading The @code{-symbol-list-variables} Command
33180 @findex -symbol-list-variables
33182 @subsubheading Synopsis
33185 -symbol-list-variables
33188 List all the global and static variable names.
33190 @subsubheading @value{GDBN} Command
33192 @samp{info variables} in @value{GDBN}, @samp{gdb_search} in @code{gdbtk}.
33194 @subsubheading Example
33198 @subheading The @code{-symbol-locate} Command
33199 @findex -symbol-locate
33201 @subsubheading Synopsis
33207 @subsubheading @value{GDBN} Command
33209 @samp{gdb_loc} in @code{gdbtk}.
33211 @subsubheading Example
33215 @subheading The @code{-symbol-type} Command
33216 @findex -symbol-type
33218 @subsubheading Synopsis
33221 -symbol-type @var{variable}
33224 Show type of @var{variable}.
33226 @subsubheading @value{GDBN} Command
33228 The corresponding @value{GDBN} command is @samp{ptype}, @code{gdbtk} has
33229 @samp{gdb_obj_variable}.
33231 @subsubheading Example
33236 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
33237 @node GDB/MI File Commands
33238 @section @sc{gdb/mi} File Commands
33240 This section describes the GDB/MI commands to specify executable file names
33241 and to read in and obtain symbol table information.
33243 @subheading The @code{-file-exec-and-symbols} Command
33244 @findex -file-exec-and-symbols
33246 @subsubheading Synopsis
33249 -file-exec-and-symbols @var{file}
33252 Specify the executable file to be debugged. This file is the one from
33253 which the symbol table is also read. If no file is specified, the
33254 command clears the executable and symbol information. If breakpoints
33255 are set when using this command with no arguments, @value{GDBN} will produce
33256 error messages. Otherwise, no output is produced, except a completion
33259 @subsubheading @value{GDBN} Command
33261 The corresponding @value{GDBN} command is @samp{file}.
33263 @subsubheading Example
33267 -file-exec-and-symbols /kwikemart/marge/ezannoni/TRUNK/mbx/hello.mbx
33273 @subheading The @code{-file-exec-file} Command
33274 @findex -file-exec-file
33276 @subsubheading Synopsis
33279 -file-exec-file @var{file}
33282 Specify the executable file to be debugged. Unlike
33283 @samp{-file-exec-and-symbols}, the symbol table is @emph{not} read
33284 from this file. If used without argument, @value{GDBN} clears the information
33285 about the executable file. No output is produced, except a completion
33288 @subsubheading @value{GDBN} Command
33290 The corresponding @value{GDBN} command is @samp{exec-file}.
33292 @subsubheading Example
33296 -file-exec-file /kwikemart/marge/ezannoni/TRUNK/mbx/hello.mbx
33303 @subheading The @code{-file-list-exec-sections} Command
33304 @findex -file-list-exec-sections
33306 @subsubheading Synopsis
33309 -file-list-exec-sections
33312 List the sections of the current executable file.
33314 @subsubheading @value{GDBN} Command
33316 The @value{GDBN} command @samp{info file} shows, among the rest, the same
33317 information as this command. @code{gdbtk} has a corresponding command
33318 @samp{gdb_load_info}.
33320 @subsubheading Example
33325 @subheading The @code{-file-list-exec-source-file} Command
33326 @findex -file-list-exec-source-file
33328 @subsubheading Synopsis
33331 -file-list-exec-source-file
33334 List the line number, the current source file, and the absolute path
33335 to the current source file for the current executable. The macro
33336 information field has a value of @samp{1} or @samp{0} depending on
33337 whether or not the file includes preprocessor macro information.
33339 @subsubheading @value{GDBN} Command
33341 The @value{GDBN} equivalent is @samp{info source}
33343 @subsubheading Example
33347 123-file-list-exec-source-file
33348 123^done,line="1",file="foo.c",fullname="/home/bar/foo.c,macro-info="1"
33353 @subheading The @code{-file-list-exec-source-files} Command
33354 @findex -file-list-exec-source-files
33356 @subsubheading Synopsis
33359 -file-list-exec-source-files
33362 List the source files for the current executable.
33364 It will always output both the filename and fullname (absolute file
33365 name) of a source file.
33367 @subsubheading @value{GDBN} Command
33369 The @value{GDBN} equivalent is @samp{info sources}.
33370 @code{gdbtk} has an analogous command @samp{gdb_listfiles}.
33372 @subsubheading Example
33375 -file-list-exec-source-files
33377 @{file=foo.c,fullname=/home/foo.c@},
33378 @{file=/home/bar.c,fullname=/home/bar.c@},
33379 @{file=gdb_could_not_find_fullpath.c@}]
33383 @subheading The @code{-file-list-shared-libraries} Command
33384 @findex -file-list-shared-libraries
33386 @subsubheading Synopsis
33389 -file-list-shared-libraries [ @var{regexp} ]
33392 List the shared libraries in the program.
33393 With a regular expression @var{regexp}, only those libraries whose
33394 names match @var{regexp} are listed.
33396 @subsubheading @value{GDBN} Command
33398 The corresponding @value{GDBN} command is @samp{info shared}. The fields
33399 have a similar meaning to the @code{=library-loaded} notification.
33400 The @code{ranges} field specifies the multiple segments belonging to this
33401 library. Each range has the following fields:
33405 The address defining the inclusive lower bound of the segment.
33407 The address defining the exclusive upper bound of the segment.
33410 @subsubheading Example
33413 -file-list-exec-source-files
33414 ^done,shared-libraries=[
33415 @{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"@}]@},
33416 @{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"@}]@}]
33422 @subheading The @code{-file-list-symbol-files} Command
33423 @findex -file-list-symbol-files
33425 @subsubheading Synopsis
33428 -file-list-symbol-files
33433 @subsubheading @value{GDBN} Command
33435 The corresponding @value{GDBN} command is @samp{info file} (part of it).
33437 @subsubheading Example
33442 @subheading The @code{-file-symbol-file} Command
33443 @findex -file-symbol-file
33445 @subsubheading Synopsis
33448 -file-symbol-file @var{file}
33451 Read symbol table info from the specified @var{file} argument. When
33452 used without arguments, clears @value{GDBN}'s symbol table info. No output is
33453 produced, except for a completion notification.
33455 @subsubheading @value{GDBN} Command
33457 The corresponding @value{GDBN} command is @samp{symbol-file}.
33459 @subsubheading Example
33463 -file-symbol-file /kwikemart/marge/ezannoni/TRUNK/mbx/hello.mbx
33469 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
33470 @node GDB/MI Memory Overlay Commands
33471 @section @sc{gdb/mi} Memory Overlay Commands
33473 The memory overlay commands are not implemented.
33475 @c @subheading -overlay-auto
33477 @c @subheading -overlay-list-mapping-state
33479 @c @subheading -overlay-list-overlays
33481 @c @subheading -overlay-map
33483 @c @subheading -overlay-off
33485 @c @subheading -overlay-on
33487 @c @subheading -overlay-unmap
33489 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
33490 @node GDB/MI Signal Handling Commands
33491 @section @sc{gdb/mi} Signal Handling Commands
33493 Signal handling commands are not implemented.
33495 @c @subheading -signal-handle
33497 @c @subheading -signal-list-handle-actions
33499 @c @subheading -signal-list-signal-types
33503 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
33504 @node GDB/MI Target Manipulation
33505 @section @sc{gdb/mi} Target Manipulation Commands
33508 @subheading The @code{-target-attach} Command
33509 @findex -target-attach
33511 @subsubheading Synopsis
33514 -target-attach @var{pid} | @var{gid} | @var{file}
33517 Attach to a process @var{pid} or a file @var{file} outside of
33518 @value{GDBN}, or a thread group @var{gid}. If attaching to a thread
33519 group, the id previously returned by
33520 @samp{-list-thread-groups --available} must be used.
33522 @subsubheading @value{GDBN} Command
33524 The corresponding @value{GDBN} command is @samp{attach}.
33526 @subsubheading Example
33530 =thread-created,id="1"
33531 *stopped,thread-id="1",frame=@{addr="0xb7f7e410",func="bar",args=[]@}
33537 @subheading The @code{-target-compare-sections} Command
33538 @findex -target-compare-sections
33540 @subsubheading Synopsis
33543 -target-compare-sections [ @var{section} ]
33546 Compare data of section @var{section} on target to the exec file.
33547 Without the argument, all sections are compared.
33549 @subsubheading @value{GDBN} Command
33551 The @value{GDBN} equivalent is @samp{compare-sections}.
33553 @subsubheading Example
33558 @subheading The @code{-target-detach} Command
33559 @findex -target-detach
33561 @subsubheading Synopsis
33564 -target-detach [ @var{pid} | @var{gid} ]
33567 Detach from the remote target which normally resumes its execution.
33568 If either @var{pid} or @var{gid} is specified, detaches from either
33569 the specified process, or specified thread group. There's no output.
33571 @subsubheading @value{GDBN} Command
33573 The corresponding @value{GDBN} command is @samp{detach}.
33575 @subsubheading Example
33585 @subheading The @code{-target-disconnect} Command
33586 @findex -target-disconnect
33588 @subsubheading Synopsis
33594 Disconnect from the remote target. There's no output and the target is
33595 generally not resumed.
33597 @subsubheading @value{GDBN} Command
33599 The corresponding @value{GDBN} command is @samp{disconnect}.
33601 @subsubheading Example
33611 @subheading The @code{-target-download} Command
33612 @findex -target-download
33614 @subsubheading Synopsis
33620 Loads the executable onto the remote target.
33621 It prints out an update message every half second, which includes the fields:
33625 The name of the section.
33627 The size of what has been sent so far for that section.
33629 The size of the section.
33631 The total size of what was sent so far (the current and the previous sections).
33633 The size of the overall executable to download.
33637 Each message is sent as status record (@pxref{GDB/MI Output Syntax, ,
33638 @sc{gdb/mi} Output Syntax}).
33640 In addition, it prints the name and size of the sections, as they are
33641 downloaded. These messages include the following fields:
33645 The name of the section.
33647 The size of the section.
33649 The size of the overall executable to download.
33653 At the end, a summary is printed.
33655 @subsubheading @value{GDBN} Command
33657 The corresponding @value{GDBN} command is @samp{load}.
33659 @subsubheading Example
33661 Note: each status message appears on a single line. Here the messages
33662 have been broken down so that they can fit onto a page.
33667 +download,@{section=".text",section-size="6668",total-size="9880"@}
33668 +download,@{section=".text",section-sent="512",section-size="6668",
33669 total-sent="512",total-size="9880"@}
33670 +download,@{section=".text",section-sent="1024",section-size="6668",
33671 total-sent="1024",total-size="9880"@}
33672 +download,@{section=".text",section-sent="1536",section-size="6668",
33673 total-sent="1536",total-size="9880"@}
33674 +download,@{section=".text",section-sent="2048",section-size="6668",
33675 total-sent="2048",total-size="9880"@}
33676 +download,@{section=".text",section-sent="2560",section-size="6668",
33677 total-sent="2560",total-size="9880"@}
33678 +download,@{section=".text",section-sent="3072",section-size="6668",
33679 total-sent="3072",total-size="9880"@}
33680 +download,@{section=".text",section-sent="3584",section-size="6668",
33681 total-sent="3584",total-size="9880"@}
33682 +download,@{section=".text",section-sent="4096",section-size="6668",
33683 total-sent="4096",total-size="9880"@}
33684 +download,@{section=".text",section-sent="4608",section-size="6668",
33685 total-sent="4608",total-size="9880"@}
33686 +download,@{section=".text",section-sent="5120",section-size="6668",
33687 total-sent="5120",total-size="9880"@}
33688 +download,@{section=".text",section-sent="5632",section-size="6668",
33689 total-sent="5632",total-size="9880"@}
33690 +download,@{section=".text",section-sent="6144",section-size="6668",
33691 total-sent="6144",total-size="9880"@}
33692 +download,@{section=".text",section-sent="6656",section-size="6668",
33693 total-sent="6656",total-size="9880"@}
33694 +download,@{section=".init",section-size="28",total-size="9880"@}
33695 +download,@{section=".fini",section-size="28",total-size="9880"@}
33696 +download,@{section=".data",section-size="3156",total-size="9880"@}
33697 +download,@{section=".data",section-sent="512",section-size="3156",
33698 total-sent="7236",total-size="9880"@}
33699 +download,@{section=".data",section-sent="1024",section-size="3156",
33700 total-sent="7748",total-size="9880"@}
33701 +download,@{section=".data",section-sent="1536",section-size="3156",
33702 total-sent="8260",total-size="9880"@}
33703 +download,@{section=".data",section-sent="2048",section-size="3156",
33704 total-sent="8772",total-size="9880"@}
33705 +download,@{section=".data",section-sent="2560",section-size="3156",
33706 total-sent="9284",total-size="9880"@}
33707 +download,@{section=".data",section-sent="3072",section-size="3156",
33708 total-sent="9796",total-size="9880"@}
33709 ^done,address="0x10004",load-size="9880",transfer-rate="6586",
33716 @subheading The @code{-target-exec-status} Command
33717 @findex -target-exec-status
33719 @subsubheading Synopsis
33722 -target-exec-status
33725 Provide information on the state of the target (whether it is running or
33726 not, for instance).
33728 @subsubheading @value{GDBN} Command
33730 There's no equivalent @value{GDBN} command.
33732 @subsubheading Example
33736 @subheading The @code{-target-list-available-targets} Command
33737 @findex -target-list-available-targets
33739 @subsubheading Synopsis
33742 -target-list-available-targets
33745 List the possible targets to connect to.
33747 @subsubheading @value{GDBN} Command
33749 The corresponding @value{GDBN} command is @samp{help target}.
33751 @subsubheading Example
33755 @subheading The @code{-target-list-current-targets} Command
33756 @findex -target-list-current-targets
33758 @subsubheading Synopsis
33761 -target-list-current-targets
33764 Describe the current target.
33766 @subsubheading @value{GDBN} Command
33768 The corresponding information is printed by @samp{info file} (among
33771 @subsubheading Example
33775 @subheading The @code{-target-list-parameters} Command
33776 @findex -target-list-parameters
33778 @subsubheading Synopsis
33781 -target-list-parameters
33787 @subsubheading @value{GDBN} Command
33791 @subsubheading Example
33794 @subheading The @code{-target-flash-erase} Command
33795 @findex -target-flash-erase
33797 @subsubheading Synopsis
33800 -target-flash-erase
33803 Erases all known flash memory regions on the target.
33805 The corresponding @value{GDBN} command is @samp{flash-erase}.
33807 The output is a list of flash regions that have been erased, with starting
33808 addresses and memory region sizes.
33812 -target-flash-erase
33813 ^done,erased-regions=@{address="0x0",size="0x40000"@}
33817 @subheading The @code{-target-select} Command
33818 @findex -target-select
33820 @subsubheading Synopsis
33823 -target-select @var{type} @var{parameters @dots{}}
33826 Connect @value{GDBN} to the remote target. This command takes two args:
33830 The type of target, for instance @samp{remote}, etc.
33831 @item @var{parameters}
33832 Device names, host names and the like. @xref{Target Commands, ,
33833 Commands for Managing Targets}, for more details.
33836 The output is a connection notification, followed by the address at
33837 which the target program is, in the following form:
33840 ^connected,addr="@var{address}",func="@var{function name}",
33841 args=[@var{arg list}]
33844 @subsubheading @value{GDBN} Command
33846 The corresponding @value{GDBN} command is @samp{target}.
33848 @subsubheading Example
33852 -target-select remote /dev/ttya
33853 ^connected,addr="0xfe00a300",func="??",args=[]
33857 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
33858 @node GDB/MI File Transfer Commands
33859 @section @sc{gdb/mi} File Transfer Commands
33862 @subheading The @code{-target-file-put} Command
33863 @findex -target-file-put
33865 @subsubheading Synopsis
33868 -target-file-put @var{hostfile} @var{targetfile}
33871 Copy file @var{hostfile} from the host system (the machine running
33872 @value{GDBN}) to @var{targetfile} on the target system.
33874 @subsubheading @value{GDBN} Command
33876 The corresponding @value{GDBN} command is @samp{remote put}.
33878 @subsubheading Example
33882 -target-file-put localfile remotefile
33888 @subheading The @code{-target-file-get} Command
33889 @findex -target-file-get
33891 @subsubheading Synopsis
33894 -target-file-get @var{targetfile} @var{hostfile}
33897 Copy file @var{targetfile} from the target system to @var{hostfile}
33898 on the host system.
33900 @subsubheading @value{GDBN} Command
33902 The corresponding @value{GDBN} command is @samp{remote get}.
33904 @subsubheading Example
33908 -target-file-get remotefile localfile
33914 @subheading The @code{-target-file-delete} Command
33915 @findex -target-file-delete
33917 @subsubheading Synopsis
33920 -target-file-delete @var{targetfile}
33923 Delete @var{targetfile} from the target system.
33925 @subsubheading @value{GDBN} Command
33927 The corresponding @value{GDBN} command is @samp{remote delete}.
33929 @subsubheading Example
33933 -target-file-delete remotefile
33939 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
33940 @node GDB/MI Ada Exceptions Commands
33941 @section Ada Exceptions @sc{gdb/mi} Commands
33943 @subheading The @code{-info-ada-exceptions} Command
33944 @findex -info-ada-exceptions
33946 @subsubheading Synopsis
33949 -info-ada-exceptions [ @var{regexp}]
33952 List all Ada exceptions defined within the program being debugged.
33953 With a regular expression @var{regexp}, only those exceptions whose
33954 names match @var{regexp} are listed.
33956 @subsubheading @value{GDBN} Command
33958 The corresponding @value{GDBN} command is @samp{info exceptions}.
33960 @subsubheading Result
33962 The result is a table of Ada exceptions. The following columns are
33963 defined for each exception:
33967 The name of the exception.
33970 The address of the exception.
33974 @subsubheading Example
33977 -info-ada-exceptions aint
33978 ^done,ada-exceptions=@{nr_rows="2",nr_cols="2",
33979 hdr=[@{width="1",alignment="-1",col_name="name",colhdr="Name"@},
33980 @{width="1",alignment="-1",col_name="address",colhdr="Address"@}],
33981 body=[@{name="constraint_error",address="0x0000000000613da0"@},
33982 @{name="const.aint_global_e",address="0x0000000000613b00"@}]@}
33985 @subheading Catching Ada Exceptions
33987 The commands describing how to ask @value{GDBN} to stop when a program
33988 raises an exception are described at @ref{Ada Exception GDB/MI
33989 Catchpoint Commands}.
33992 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
33993 @node GDB/MI Support Commands
33994 @section @sc{gdb/mi} Support Commands
33996 Since new commands and features get regularly added to @sc{gdb/mi},
33997 some commands are available to help front-ends query the debugger
33998 about support for these capabilities. Similarly, it is also possible
33999 to query @value{GDBN} about target support of certain features.
34001 @subheading The @code{-info-gdb-mi-command} Command
34002 @cindex @code{-info-gdb-mi-command}
34003 @findex -info-gdb-mi-command
34005 @subsubheading Synopsis
34008 -info-gdb-mi-command @var{cmd_name}
34011 Query support for the @sc{gdb/mi} command named @var{cmd_name}.
34013 Note that the dash (@code{-}) starting all @sc{gdb/mi} commands
34014 is technically not part of the command name (@pxref{GDB/MI Input
34015 Syntax}), and thus should be omitted in @var{cmd_name}. However,
34016 for ease of use, this command also accepts the form with the leading
34019 @subsubheading @value{GDBN} Command
34021 There is no corresponding @value{GDBN} command.
34023 @subsubheading Result
34025 The result is a tuple. There is currently only one field:
34029 This field is equal to @code{"true"} if the @sc{gdb/mi} command exists,
34030 @code{"false"} otherwise.
34034 @subsubheading Example
34036 Here is an example where the @sc{gdb/mi} command does not exist:
34039 -info-gdb-mi-command unsupported-command
34040 ^done,command=@{exists="false"@}
34044 And here is an example where the @sc{gdb/mi} command is known
34048 -info-gdb-mi-command symbol-list-lines
34049 ^done,command=@{exists="true"@}
34052 @subheading The @code{-list-features} Command
34053 @findex -list-features
34054 @cindex supported @sc{gdb/mi} features, list
34056 Returns a list of particular features of the MI protocol that
34057 this version of gdb implements. A feature can be a command,
34058 or a new field in an output of some command, or even an
34059 important bugfix. While a frontend can sometimes detect presence
34060 of a feature at runtime, it is easier to perform detection at debugger
34063 The command returns a list of strings, with each string naming an
34064 available feature. Each returned string is just a name, it does not
34065 have any internal structure. The list of possible feature names
34071 (gdb) -list-features
34072 ^done,result=["feature1","feature2"]
34075 The current list of features is:
34078 @item frozen-varobjs
34079 Indicates support for the @code{-var-set-frozen} command, as well
34080 as possible presense of the @code{frozen} field in the output
34081 of @code{-varobj-create}.
34082 @item pending-breakpoints
34083 Indicates support for the @option{-f} option to the @code{-break-insert}
34086 Indicates Python scripting support, Python-based
34087 pretty-printing commands, and possible presence of the
34088 @samp{display_hint} field in the output of @code{-var-list-children}
34090 Indicates support for the @code{-thread-info} command.
34091 @item data-read-memory-bytes
34092 Indicates support for the @code{-data-read-memory-bytes} and the
34093 @code{-data-write-memory-bytes} commands.
34094 @item breakpoint-notifications
34095 Indicates that changes to breakpoints and breakpoints created via the
34096 CLI will be announced via async records.
34097 @item ada-task-info
34098 Indicates support for the @code{-ada-task-info} command.
34099 @item language-option
34100 Indicates that all @sc{gdb/mi} commands accept the @option{--language}
34101 option (@pxref{Context management}).
34102 @item info-gdb-mi-command
34103 Indicates support for the @code{-info-gdb-mi-command} command.
34104 @item undefined-command-error-code
34105 Indicates support for the "undefined-command" error code in error result
34106 records, produced when trying to execute an undefined @sc{gdb/mi} command
34107 (@pxref{GDB/MI Result Records}).
34108 @item exec-run-start-option
34109 Indicates that the @code{-exec-run} command supports the @option{--start}
34110 option (@pxref{GDB/MI Program Execution}).
34111 @item data-disassemble-a-option
34112 Indicates that the @code{-data-disassemble} command supports the @option{-a}
34113 option (@pxref{GDB/MI Data Manipulation}).
34116 @subheading The @code{-list-target-features} Command
34117 @findex -list-target-features
34119 Returns a list of particular features that are supported by the
34120 target. Those features affect the permitted MI commands, but
34121 unlike the features reported by the @code{-list-features} command, the
34122 features depend on which target GDB is using at the moment. Whenever
34123 a target can change, due to commands such as @code{-target-select},
34124 @code{-target-attach} or @code{-exec-run}, the list of target features
34125 may change, and the frontend should obtain it again.
34129 (gdb) -list-target-features
34130 ^done,result=["async"]
34133 The current list of features is:
34137 Indicates that the target is capable of asynchronous command
34138 execution, which means that @value{GDBN} will accept further commands
34139 while the target is running.
34142 Indicates that the target is capable of reverse execution.
34143 @xref{Reverse Execution}, for more information.
34147 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
34148 @node GDB/MI Miscellaneous Commands
34149 @section Miscellaneous @sc{gdb/mi} Commands
34151 @c @subheading -gdb-complete
34153 @subheading The @code{-gdb-exit} Command
34156 @subsubheading Synopsis
34162 Exit @value{GDBN} immediately.
34164 @subsubheading @value{GDBN} Command
34166 Approximately corresponds to @samp{quit}.
34168 @subsubheading Example
34178 @subheading The @code{-exec-abort} Command
34179 @findex -exec-abort
34181 @subsubheading Synopsis
34187 Kill the inferior running program.
34189 @subsubheading @value{GDBN} Command
34191 The corresponding @value{GDBN} command is @samp{kill}.
34193 @subsubheading Example
34198 @subheading The @code{-gdb-set} Command
34201 @subsubheading Synopsis
34207 Set an internal @value{GDBN} variable.
34208 @c IS THIS A DOLLAR VARIABLE? OR SOMETHING LIKE ANNOTATE ?????
34210 @subsubheading @value{GDBN} Command
34212 The corresponding @value{GDBN} command is @samp{set}.
34214 @subsubheading Example
34224 @subheading The @code{-gdb-show} Command
34227 @subsubheading Synopsis
34233 Show the current value of a @value{GDBN} variable.
34235 @subsubheading @value{GDBN} Command
34237 The corresponding @value{GDBN} command is @samp{show}.
34239 @subsubheading Example
34248 @c @subheading -gdb-source
34251 @subheading The @code{-gdb-version} Command
34252 @findex -gdb-version
34254 @subsubheading Synopsis
34260 Show version information for @value{GDBN}. Used mostly in testing.
34262 @subsubheading @value{GDBN} Command
34264 The @value{GDBN} equivalent is @samp{show version}. @value{GDBN} by
34265 default shows this information when you start an interactive session.
34267 @subsubheading Example
34269 @c This example modifies the actual output from GDB to avoid overfull
34275 ~Copyright 2000 Free Software Foundation, Inc.
34276 ~GDB is free software, covered by the GNU General Public License, and
34277 ~you are welcome to change it and/or distribute copies of it under
34278 ~ certain conditions.
34279 ~Type "show copying" to see the conditions.
34280 ~There is absolutely no warranty for GDB. Type "show warranty" for
34282 ~This GDB was configured as
34283 "--host=sparc-sun-solaris2.5.1 --target=ppc-eabi".
34288 @subheading The @code{-list-thread-groups} Command
34289 @findex -list-thread-groups
34291 @subheading Synopsis
34294 -list-thread-groups [ --available ] [ --recurse 1 ] [ @var{group} ... ]
34297 Lists thread groups (@pxref{Thread groups}). When a single thread
34298 group is passed as the argument, lists the children of that group.
34299 When several thread group are passed, lists information about those
34300 thread groups. Without any parameters, lists information about all
34301 top-level thread groups.
34303 Normally, thread groups that are being debugged are reported.
34304 With the @samp{--available} option, @value{GDBN} reports thread groups
34305 available on the target.
34307 The output of this command may have either a @samp{threads} result or
34308 a @samp{groups} result. The @samp{thread} result has a list of tuples
34309 as value, with each tuple describing a thread (@pxref{GDB/MI Thread
34310 Information}). The @samp{groups} result has a list of tuples as value,
34311 each tuple describing a thread group. If top-level groups are
34312 requested (that is, no parameter is passed), or when several groups
34313 are passed, the output always has a @samp{groups} result. The format
34314 of the @samp{group} result is described below.
34316 To reduce the number of roundtrips it's possible to list thread groups
34317 together with their children, by passing the @samp{--recurse} option
34318 and the recursion depth. Presently, only recursion depth of 1 is
34319 permitted. If this option is present, then every reported thread group
34320 will also include its children, either as @samp{group} or
34321 @samp{threads} field.
34323 In general, any combination of option and parameters is permitted, with
34324 the following caveats:
34328 When a single thread group is passed, the output will typically
34329 be the @samp{threads} result. Because threads may not contain
34330 anything, the @samp{recurse} option will be ignored.
34333 When the @samp{--available} option is passed, limited information may
34334 be available. In particular, the list of threads of a process might
34335 be inaccessible. Further, specifying specific thread groups might
34336 not give any performance advantage over listing all thread groups.
34337 The frontend should assume that @samp{-list-thread-groups --available}
34338 is always an expensive operation and cache the results.
34342 The @samp{groups} result is a list of tuples, where each tuple may
34343 have the following fields:
34347 Identifier of the thread group. This field is always present.
34348 The identifier is an opaque string; frontends should not try to
34349 convert it to an integer, even though it might look like one.
34352 The type of the thread group. At present, only @samp{process} is a
34356 The target-specific process identifier. This field is only present
34357 for thread groups of type @samp{process} and only if the process exists.
34360 The exit code of this group's last exited thread, formatted in octal.
34361 This field is only present for thread groups of type @samp{process} and
34362 only if the process is not running.
34365 The number of children this thread group has. This field may be
34366 absent for an available thread group.
34369 This field has a list of tuples as value, each tuple describing a
34370 thread. It may be present if the @samp{--recurse} option is
34371 specified, and it's actually possible to obtain the threads.
34374 This field is a list of integers, each identifying a core that one
34375 thread of the group is running on. This field may be absent if
34376 such information is not available.
34379 The name of the executable file that corresponds to this thread group.
34380 The field is only present for thread groups of type @samp{process},
34381 and only if there is a corresponding executable file.
34385 @subheading Example
34389 -list-thread-groups
34390 ^done,groups=[@{id="17",type="process",pid="yyy",num_children="2"@}]
34391 -list-thread-groups 17
34392 ^done,threads=[@{id="2",target-id="Thread 0xb7e14b90 (LWP 21257)",
34393 frame=@{level="0",addr="0xffffe410",func="__kernel_vsyscall",args=[]@},state="running"@},
34394 @{id="1",target-id="Thread 0xb7e156b0 (LWP 21254)",
34395 frame=@{level="0",addr="0x0804891f",func="foo",args=[@{name="i",value="10"@}],
34396 file="/tmp/a.c",fullname="/tmp/a.c",line="158",arch="i386:x86_64"@},state="running"@}]]
34397 -list-thread-groups --available
34398 ^done,groups=[@{id="17",type="process",pid="yyy",num_children="2",cores=[1,2]@}]
34399 -list-thread-groups --available --recurse 1
34400 ^done,groups=[@{id="17", types="process",pid="yyy",num_children="2",cores=[1,2],
34401 threads=[@{id="1",target-id="Thread 0xb7e14b90",cores=[1]@},
34402 @{id="2",target-id="Thread 0xb7e14b90",cores=[2]@}]@},..]
34403 -list-thread-groups --available --recurse 1 17 18
34404 ^done,groups=[@{id="17", types="process",pid="yyy",num_children="2",cores=[1,2],
34405 threads=[@{id="1",target-id="Thread 0xb7e14b90",cores=[1]@},
34406 @{id="2",target-id="Thread 0xb7e14b90",cores=[2]@}]@},...]
34409 @subheading The @code{-info-os} Command
34412 @subsubheading Synopsis
34415 -info-os [ @var{type} ]
34418 If no argument is supplied, the command returns a table of available
34419 operating-system-specific information types. If one of these types is
34420 supplied as an argument @var{type}, then the command returns a table
34421 of data of that type.
34423 The types of information available depend on the target operating
34426 @subsubheading @value{GDBN} Command
34428 The corresponding @value{GDBN} command is @samp{info os}.
34430 @subsubheading Example
34432 When run on a @sc{gnu}/Linux system, the output will look something
34438 ^done,OSDataTable=@{nr_rows="10",nr_cols="3",
34439 hdr=[@{width="10",alignment="-1",col_name="col0",colhdr="Type"@},
34440 @{width="10",alignment="-1",col_name="col1",colhdr="Description"@},
34441 @{width="10",alignment="-1",col_name="col2",colhdr="Title"@}],
34442 body=[item=@{col0="cpus",col1="Listing of all cpus/cores on the system",
34444 item=@{col0="files",col1="Listing of all file descriptors",
34445 col2="File descriptors"@},
34446 item=@{col0="modules",col1="Listing of all loaded kernel modules",
34447 col2="Kernel modules"@},
34448 item=@{col0="msg",col1="Listing of all message queues",
34449 col2="Message queues"@},
34450 item=@{col0="processes",col1="Listing of all processes",
34451 col2="Processes"@},
34452 item=@{col0="procgroups",col1="Listing of all process groups",
34453 col2="Process groups"@},
34454 item=@{col0="semaphores",col1="Listing of all semaphores",
34455 col2="Semaphores"@},
34456 item=@{col0="shm",col1="Listing of all shared-memory regions",
34457 col2="Shared-memory regions"@},
34458 item=@{col0="sockets",col1="Listing of all internet-domain sockets",
34460 item=@{col0="threads",col1="Listing of all threads",
34464 ^done,OSDataTable=@{nr_rows="190",nr_cols="4",
34465 hdr=[@{width="10",alignment="-1",col_name="col0",colhdr="pid"@},
34466 @{width="10",alignment="-1",col_name="col1",colhdr="user"@},
34467 @{width="10",alignment="-1",col_name="col2",colhdr="command"@},
34468 @{width="10",alignment="-1",col_name="col3",colhdr="cores"@}],
34469 body=[item=@{col0="1",col1="root",col2="/sbin/init",col3="0"@},
34470 item=@{col0="2",col1="root",col2="[kthreadd]",col3="1"@},
34471 item=@{col0="3",col1="root",col2="[ksoftirqd/0]",col3="0"@},
34473 item=@{col0="26446",col1="stan",col2="bash",col3="0"@},
34474 item=@{col0="28152",col1="stan",col2="bash",col3="1"@}]@}
34478 (Note that the MI output here includes a @code{"Title"} column that
34479 does not appear in command-line @code{info os}; this column is useful
34480 for MI clients that want to enumerate the types of data, such as in a
34481 popup menu, but is needless clutter on the command line, and
34482 @code{info os} omits it.)
34484 @subheading The @code{-add-inferior} Command
34485 @findex -add-inferior
34487 @subheading Synopsis
34493 Creates a new inferior (@pxref{Inferiors and Programs}). The created
34494 inferior is not associated with any executable. Such association may
34495 be established with the @samp{-file-exec-and-symbols} command
34496 (@pxref{GDB/MI File Commands}). The command response has a single
34497 field, @samp{inferior}, whose value is the identifier of the
34498 thread group corresponding to the new inferior.
34500 @subheading Example
34505 ^done,inferior="i3"
34508 @subheading The @code{-interpreter-exec} Command
34509 @findex -interpreter-exec
34511 @subheading Synopsis
34514 -interpreter-exec @var{interpreter} @var{command}
34516 @anchor{-interpreter-exec}
34518 Execute the specified @var{command} in the given @var{interpreter}.
34520 @subheading @value{GDBN} Command
34522 The corresponding @value{GDBN} command is @samp{interpreter-exec}.
34524 @subheading Example
34528 -interpreter-exec console "break main"
34529 &"During symbol reading, couldn't parse type; debugger out of date?.\n"
34530 &"During symbol reading, bad structure-type format.\n"
34531 ~"Breakpoint 1 at 0x8074fc6: file ../../src/gdb/main.c, line 743.\n"
34536 @subheading The @code{-inferior-tty-set} Command
34537 @findex -inferior-tty-set
34539 @subheading Synopsis
34542 -inferior-tty-set /dev/pts/1
34545 Set terminal for future runs of the program being debugged.
34547 @subheading @value{GDBN} Command
34549 The corresponding @value{GDBN} command is @samp{set inferior-tty} /dev/pts/1.
34551 @subheading Example
34555 -inferior-tty-set /dev/pts/1
34560 @subheading The @code{-inferior-tty-show} Command
34561 @findex -inferior-tty-show
34563 @subheading Synopsis
34569 Show terminal for future runs of program being debugged.
34571 @subheading @value{GDBN} Command
34573 The corresponding @value{GDBN} command is @samp{show inferior-tty}.
34575 @subheading Example
34579 -inferior-tty-set /dev/pts/1
34583 ^done,inferior_tty_terminal="/dev/pts/1"
34587 @subheading The @code{-enable-timings} Command
34588 @findex -enable-timings
34590 @subheading Synopsis
34593 -enable-timings [yes | no]
34596 Toggle the printing of the wallclock, user and system times for an MI
34597 command as a field in its output. This command is to help frontend
34598 developers optimize the performance of their code. No argument is
34599 equivalent to @samp{yes}.
34601 @subheading @value{GDBN} Command
34605 @subheading Example
34613 ^done,bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
34614 addr="0x080484ed",func="main",file="myprog.c",
34615 fullname="/home/nickrob/myprog.c",line="73",thread-groups=["i1"],
34617 time=@{wallclock="0.05185",user="0.00800",system="0.00000"@}
34625 *stopped,reason="breakpoint-hit",disp="keep",bkptno="1",thread-id="0",
34626 frame=@{addr="0x080484ed",func="main",args=[@{name="argc",value="1"@},
34627 @{name="argv",value="0xbfb60364"@}],file="myprog.c",
34628 fullname="/home/nickrob/myprog.c",line="73",arch="i386:x86_64"@}
34632 @subheading The @code{-complete} Command
34635 @subheading Synopsis
34638 -complete @var{command}
34641 Show a list of completions for partially typed CLI @var{command}.
34643 This command is intended for @sc{gdb/mi} frontends that cannot use two separate
34644 CLI and MI channels --- for example: because of lack of PTYs like on Windows or
34645 because @value{GDBN} is used remotely via a SSH connection.
34649 The result consists of two or three fields:
34653 This field contains the completed @var{command}. If @var{command}
34654 has no known completions, this field is omitted.
34657 This field contains a (possibly empty) array of matches. It is always present.
34659 @item max_completions_reached
34660 This field contains @code{1} if number of known completions is above
34661 @code{max-completions} limit (@pxref{Completion}), otherwise it contains
34662 @code{0}. It is always present.
34666 @subheading @value{GDBN} Command
34668 The corresponding @value{GDBN} command is @samp{complete}.
34670 @subheading Example
34675 ^done,completion="break",
34676 matches=["break","break-range"],
34677 max_completions_reached="0"
34680 ^done,completion="b ma",
34681 matches=["b madvise","b main"],max_completions_reached="0"
34683 -complete "b push_b"
34684 ^done,completion="b push_back(",
34686 "b A::push_back(void*)",
34687 "b std::string::push_back(char)",
34688 "b std::vector<int, std::allocator<int> >::push_back(int&&)"],
34689 max_completions_reached="0"
34691 -complete "nonexist"
34692 ^done,matches=[],max_completions_reached="0"
34698 @chapter @value{GDBN} Annotations
34700 This chapter describes annotations in @value{GDBN}. Annotations were
34701 designed to interface @value{GDBN} to graphical user interfaces or other
34702 similar programs which want to interact with @value{GDBN} at a
34703 relatively high level.
34705 The annotation mechanism has largely been superseded by @sc{gdb/mi}
34709 This is Edition @value{EDITION}, @value{DATE}.
34713 * Annotations Overview:: What annotations are; the general syntax.
34714 * Server Prefix:: Issuing a command without affecting user state.
34715 * Prompting:: Annotations marking @value{GDBN}'s need for input.
34716 * Errors:: Annotations for error messages.
34717 * Invalidation:: Some annotations describe things now invalid.
34718 * Annotations for Running::
34719 Whether the program is running, how it stopped, etc.
34720 * Source Annotations:: Annotations describing source code.
34723 @node Annotations Overview
34724 @section What is an Annotation?
34725 @cindex annotations
34727 Annotations start with a newline character, two @samp{control-z}
34728 characters, and the name of the annotation. If there is no additional
34729 information associated with this annotation, the name of the annotation
34730 is followed immediately by a newline. If there is additional
34731 information, the name of the annotation is followed by a space, the
34732 additional information, and a newline. The additional information
34733 cannot contain newline characters.
34735 Any output not beginning with a newline and two @samp{control-z}
34736 characters denotes literal output from @value{GDBN}. Currently there is
34737 no need for @value{GDBN} to output a newline followed by two
34738 @samp{control-z} characters, but if there was such a need, the
34739 annotations could be extended with an @samp{escape} annotation which
34740 means those three characters as output.
34742 The annotation @var{level}, which is specified using the
34743 @option{--annotate} command line option (@pxref{Mode Options}), controls
34744 how much information @value{GDBN} prints together with its prompt,
34745 values of expressions, source lines, and other types of output. Level 0
34746 is for no annotations, level 1 is for use when @value{GDBN} is run as a
34747 subprocess of @sc{gnu} Emacs, level 3 is the maximum annotation suitable
34748 for programs that control @value{GDBN}, and level 2 annotations have
34749 been made obsolete (@pxref{Limitations, , Limitations of the Annotation
34750 Interface, annotate, GDB's Obsolete Annotations}).
34753 @kindex set annotate
34754 @item set annotate @var{level}
34755 The @value{GDBN} command @code{set annotate} sets the level of
34756 annotations to the specified @var{level}.
34758 @item show annotate
34759 @kindex show annotate
34760 Show the current annotation level.
34763 This chapter describes level 3 annotations.
34765 A simple example of starting up @value{GDBN} with annotations is:
34768 $ @kbd{gdb --annotate=3}
34770 Copyright 2003 Free Software Foundation, Inc.
34771 GDB is free software, covered by the GNU General Public License,
34772 and you are welcome to change it and/or distribute copies of it
34773 under certain conditions.
34774 Type "show copying" to see the conditions.
34775 There is absolutely no warranty for GDB. Type "show warranty"
34777 This GDB was configured as "i386-pc-linux-gnu"
34788 Here @samp{quit} is input to @value{GDBN}; the rest is output from
34789 @value{GDBN}. The three lines beginning @samp{^Z^Z} (where @samp{^Z}
34790 denotes a @samp{control-z} character) are annotations; the rest is
34791 output from @value{GDBN}.
34793 @node Server Prefix
34794 @section The Server Prefix
34795 @cindex server prefix
34797 If you prefix a command with @samp{server } then it will not affect
34798 the command history, nor will it affect @value{GDBN}'s notion of which
34799 command to repeat if @key{RET} is pressed on a line by itself. This
34800 means that commands can be run behind a user's back by a front-end in
34801 a transparent manner.
34803 The @code{server } prefix does not affect the recording of values into
34804 the value history; to print a value without recording it into the
34805 value history, use the @code{output} command instead of the
34806 @code{print} command.
34808 Using this prefix also disables confirmation requests
34809 (@pxref{confirmation requests}).
34812 @section Annotation for @value{GDBN} Input
34814 @cindex annotations for prompts
34815 When @value{GDBN} prompts for input, it annotates this fact so it is possible
34816 to know when to send output, when the output from a given command is
34819 Different kinds of input each have a different @dfn{input type}. Each
34820 input type has three annotations: a @code{pre-} annotation, which
34821 denotes the beginning of any prompt which is being output, a plain
34822 annotation, which denotes the end of the prompt, and then a @code{post-}
34823 annotation which denotes the end of any echo which may (or may not) be
34824 associated with the input. For example, the @code{prompt} input type
34825 features the following annotations:
34833 The input types are
34836 @findex pre-prompt annotation
34837 @findex prompt annotation
34838 @findex post-prompt annotation
34840 When @value{GDBN} is prompting for a command (the main @value{GDBN} prompt).
34842 @findex pre-commands annotation
34843 @findex commands annotation
34844 @findex post-commands annotation
34846 When @value{GDBN} prompts for a set of commands, like in the @code{commands}
34847 command. The annotations are repeated for each command which is input.
34849 @findex pre-overload-choice annotation
34850 @findex overload-choice annotation
34851 @findex post-overload-choice annotation
34852 @item overload-choice
34853 When @value{GDBN} wants the user to select between various overloaded functions.
34855 @findex pre-query annotation
34856 @findex query annotation
34857 @findex post-query annotation
34859 When @value{GDBN} wants the user to confirm a potentially dangerous operation.
34861 @findex pre-prompt-for-continue annotation
34862 @findex prompt-for-continue annotation
34863 @findex post-prompt-for-continue annotation
34864 @item prompt-for-continue
34865 When @value{GDBN} is asking the user to press return to continue. Note: Don't
34866 expect this to work well; instead use @code{set height 0} to disable
34867 prompting. This is because the counting of lines is buggy in the
34868 presence of annotations.
34873 @cindex annotations for errors, warnings and interrupts
34875 @findex quit annotation
34880 This annotation occurs right before @value{GDBN} responds to an interrupt.
34882 @findex error annotation
34887 This annotation occurs right before @value{GDBN} responds to an error.
34889 Quit and error annotations indicate that any annotations which @value{GDBN} was
34890 in the middle of may end abruptly. For example, if a
34891 @code{value-history-begin} annotation is followed by a @code{error}, one
34892 cannot expect to receive the matching @code{value-history-end}. One
34893 cannot expect not to receive it either, however; an error annotation
34894 does not necessarily mean that @value{GDBN} is immediately returning all the way
34897 @findex error-begin annotation
34898 A quit or error annotation may be preceded by
34904 Any output between that and the quit or error annotation is the error
34907 Warning messages are not yet annotated.
34908 @c If we want to change that, need to fix warning(), type_error(),
34909 @c range_error(), and possibly other places.
34912 @section Invalidation Notices
34914 @cindex annotations for invalidation messages
34915 The following annotations say that certain pieces of state may have
34919 @findex frames-invalid annotation
34920 @item ^Z^Zframes-invalid
34922 The frames (for example, output from the @code{backtrace} command) may
34925 @findex breakpoints-invalid annotation
34926 @item ^Z^Zbreakpoints-invalid
34928 The breakpoints may have changed. For example, the user just added or
34929 deleted a breakpoint.
34932 @node Annotations for Running
34933 @section Running the Program
34934 @cindex annotations for running programs
34936 @findex starting annotation
34937 @findex stopping annotation
34938 When the program starts executing due to a @value{GDBN} command such as
34939 @code{step} or @code{continue},
34945 is output. When the program stops,
34951 is output. Before the @code{stopped} annotation, a variety of
34952 annotations describe how the program stopped.
34955 @findex exited annotation
34956 @item ^Z^Zexited @var{exit-status}
34957 The program exited, and @var{exit-status} is the exit status (zero for
34958 successful exit, otherwise nonzero).
34960 @findex signalled annotation
34961 @findex signal-name annotation
34962 @findex signal-name-end annotation
34963 @findex signal-string annotation
34964 @findex signal-string-end annotation
34965 @item ^Z^Zsignalled
34966 The program exited with a signal. After the @code{^Z^Zsignalled}, the
34967 annotation continues:
34973 ^Z^Zsignal-name-end
34977 ^Z^Zsignal-string-end
34982 where @var{name} is the name of the signal, such as @code{SIGILL} or
34983 @code{SIGSEGV}, and @var{string} is the explanation of the signal, such
34984 as @code{Illegal Instruction} or @code{Segmentation fault}. The arguments
34985 @var{intro-text}, @var{middle-text}, and @var{end-text} are for the
34986 user's benefit and have no particular format.
34988 @findex signal annotation
34990 The syntax of this annotation is just like @code{signalled}, but @value{GDBN} is
34991 just saying that the program received the signal, not that it was
34992 terminated with it.
34994 @findex breakpoint annotation
34995 @item ^Z^Zbreakpoint @var{number}
34996 The program hit breakpoint number @var{number}.
34998 @findex watchpoint annotation
34999 @item ^Z^Zwatchpoint @var{number}
35000 The program hit watchpoint number @var{number}.
35003 @node Source Annotations
35004 @section Displaying Source
35005 @cindex annotations for source display
35007 @findex source annotation
35008 The following annotation is used instead of displaying source code:
35011 ^Z^Zsource @var{filename}:@var{line}:@var{character}:@var{middle}:@var{addr}
35014 where @var{filename} is an absolute file name indicating which source
35015 file, @var{line} is the line number within that file (where 1 is the
35016 first line in the file), @var{character} is the character position
35017 within the file (where 0 is the first character in the file) (for most
35018 debug formats this will necessarily point to the beginning of a line),
35019 @var{middle} is @samp{middle} if @var{addr} is in the middle of the
35020 line, or @samp{beg} if @var{addr} is at the beginning of the line, and
35021 @var{addr} is the address in the target program associated with the
35022 source which is being displayed. The @var{addr} is in the form @samp{0x}
35023 followed by one or more lowercase hex digits (note that this does not
35024 depend on the language).
35026 @node JIT Interface
35027 @chapter JIT Compilation Interface
35028 @cindex just-in-time compilation
35029 @cindex JIT compilation interface
35031 This chapter documents @value{GDBN}'s @dfn{just-in-time} (JIT) compilation
35032 interface. A JIT compiler is a program or library that generates native
35033 executable code at runtime and executes it, usually in order to achieve good
35034 performance while maintaining platform independence.
35036 Programs that use JIT compilation are normally difficult to debug because
35037 portions of their code are generated at runtime, instead of being loaded from
35038 object files, which is where @value{GDBN} normally finds the program's symbols
35039 and debug information. In order to debug programs that use JIT compilation,
35040 @value{GDBN} has an interface that allows the program to register in-memory
35041 symbol files with @value{GDBN} at runtime.
35043 If you are using @value{GDBN} to debug a program that uses this interface, then
35044 it should work transparently so long as you have not stripped the binary. If
35045 you are developing a JIT compiler, then the interface is documented in the rest
35046 of this chapter. At this time, the only known client of this interface is the
35049 Broadly speaking, the JIT interface mirrors the dynamic loader interface. The
35050 JIT compiler communicates with @value{GDBN} by writing data into a global
35051 variable and calling a fuction at a well-known symbol. When @value{GDBN}
35052 attaches, it reads a linked list of symbol files from the global variable to
35053 find existing code, and puts a breakpoint in the function so that it can find
35054 out about additional code.
35057 * Declarations:: Relevant C struct declarations
35058 * Registering Code:: Steps to register code
35059 * Unregistering Code:: Steps to unregister code
35060 * Custom Debug Info:: Emit debug information in a custom format
35064 @section JIT Declarations
35066 These are the relevant struct declarations that a C program should include to
35067 implement the interface:
35077 struct jit_code_entry
35079 struct jit_code_entry *next_entry;
35080 struct jit_code_entry *prev_entry;
35081 const char *symfile_addr;
35082 uint64_t symfile_size;
35085 struct jit_descriptor
35088 /* This type should be jit_actions_t, but we use uint32_t
35089 to be explicit about the bitwidth. */
35090 uint32_t action_flag;
35091 struct jit_code_entry *relevant_entry;
35092 struct jit_code_entry *first_entry;
35095 /* GDB puts a breakpoint in this function. */
35096 void __attribute__((noinline)) __jit_debug_register_code() @{ @};
35098 /* Make sure to specify the version statically, because the
35099 debugger may check the version before we can set it. */
35100 struct jit_descriptor __jit_debug_descriptor = @{ 1, 0, 0, 0 @};
35103 If the JIT is multi-threaded, then it is important that the JIT synchronize any
35104 modifications to this global data properly, which can easily be done by putting
35105 a global mutex around modifications to these structures.
35107 @node Registering Code
35108 @section Registering Code
35110 To register code with @value{GDBN}, the JIT should follow this protocol:
35114 Generate an object file in memory with symbols and other desired debug
35115 information. The file must include the virtual addresses of the sections.
35118 Create a code entry for the file, which gives the start and size of the symbol
35122 Add it to the linked list in the JIT descriptor.
35125 Point the relevant_entry field of the descriptor at the entry.
35128 Set @code{action_flag} to @code{JIT_REGISTER} and call
35129 @code{__jit_debug_register_code}.
35132 When @value{GDBN} is attached and the breakpoint fires, @value{GDBN} uses the
35133 @code{relevant_entry} pointer so it doesn't have to walk the list looking for
35134 new code. However, the linked list must still be maintained in order to allow
35135 @value{GDBN} to attach to a running process and still find the symbol files.
35137 @node Unregistering Code
35138 @section Unregistering Code
35140 If code is freed, then the JIT should use the following protocol:
35144 Remove the code entry corresponding to the code from the linked list.
35147 Point the @code{relevant_entry} field of the descriptor at the code entry.
35150 Set @code{action_flag} to @code{JIT_UNREGISTER} and call
35151 @code{__jit_debug_register_code}.
35154 If the JIT frees or recompiles code without unregistering it, then @value{GDBN}
35155 and the JIT will leak the memory used for the associated symbol files.
35157 @node Custom Debug Info
35158 @section Custom Debug Info
35159 @cindex custom JIT debug info
35160 @cindex JIT debug info reader
35162 Generating debug information in platform-native file formats (like ELF
35163 or COFF) may be an overkill for JIT compilers; especially if all the
35164 debug info is used for is displaying a meaningful backtrace. The
35165 issue can be resolved by having the JIT writers decide on a debug info
35166 format and also provide a reader that parses the debug info generated
35167 by the JIT compiler. This section gives a brief overview on writing
35168 such a parser. More specific details can be found in the source file
35169 @file{gdb/jit-reader.in}, which is also installed as a header at
35170 @file{@var{includedir}/gdb/jit-reader.h} for easy inclusion.
35172 The reader is implemented as a shared object (so this functionality is
35173 not available on platforms which don't allow loading shared objects at
35174 runtime). Two @value{GDBN} commands, @code{jit-reader-load} and
35175 @code{jit-reader-unload} are provided, to be used to load and unload
35176 the readers from a preconfigured directory. Once loaded, the shared
35177 object is used the parse the debug information emitted by the JIT
35181 * Using JIT Debug Info Readers:: How to use supplied readers correctly
35182 * Writing JIT Debug Info Readers:: Creating a debug-info reader
35185 @node Using JIT Debug Info Readers
35186 @subsection Using JIT Debug Info Readers
35187 @kindex jit-reader-load
35188 @kindex jit-reader-unload
35190 Readers can be loaded and unloaded using the @code{jit-reader-load}
35191 and @code{jit-reader-unload} commands.
35194 @item jit-reader-load @var{reader}
35195 Load the JIT reader named @var{reader}, which is a shared
35196 object specified as either an absolute or a relative file name. In
35197 the latter case, @value{GDBN} will try to load the reader from a
35198 pre-configured directory, usually @file{@var{libdir}/gdb/} on a UNIX
35199 system (here @var{libdir} is the system library directory, often
35200 @file{/usr/local/lib}).
35202 Only one reader can be active at a time; trying to load a second
35203 reader when one is already loaded will result in @value{GDBN}
35204 reporting an error. A new JIT reader can be loaded by first unloading
35205 the current one using @code{jit-reader-unload} and then invoking
35206 @code{jit-reader-load}.
35208 @item jit-reader-unload
35209 Unload the currently loaded JIT reader.
35213 @node Writing JIT Debug Info Readers
35214 @subsection Writing JIT Debug Info Readers
35215 @cindex writing JIT debug info readers
35217 As mentioned, a reader is essentially a shared object conforming to a
35218 certain ABI. This ABI is described in @file{jit-reader.h}.
35220 @file{jit-reader.h} defines the structures, macros and functions
35221 required to write a reader. It is installed (along with
35222 @value{GDBN}), in @file{@var{includedir}/gdb} where @var{includedir} is
35223 the system include directory.
35225 Readers need to be released under a GPL compatible license. A reader
35226 can be declared as released under such a license by placing the macro
35227 @code{GDB_DECLARE_GPL_COMPATIBLE_READER} in a source file.
35229 The entry point for readers is the symbol @code{gdb_init_reader},
35230 which is expected to be a function with the prototype
35232 @findex gdb_init_reader
35234 extern struct gdb_reader_funcs *gdb_init_reader (void);
35237 @cindex @code{struct gdb_reader_funcs}
35239 @code{struct gdb_reader_funcs} contains a set of pointers to callback
35240 functions. These functions are executed to read the debug info
35241 generated by the JIT compiler (@code{read}), to unwind stack frames
35242 (@code{unwind}) and to create canonical frame IDs
35243 (@code{get_Frame_id}). It also has a callback that is called when the
35244 reader is being unloaded (@code{destroy}). The struct looks like this
35247 struct gdb_reader_funcs
35249 /* Must be set to GDB_READER_INTERFACE_VERSION. */
35250 int reader_version;
35252 /* For use by the reader. */
35255 gdb_read_debug_info *read;
35256 gdb_unwind_frame *unwind;
35257 gdb_get_frame_id *get_frame_id;
35258 gdb_destroy_reader *destroy;
35262 @cindex @code{struct gdb_symbol_callbacks}
35263 @cindex @code{struct gdb_unwind_callbacks}
35265 The callbacks are provided with another set of callbacks by
35266 @value{GDBN} to do their job. For @code{read}, these callbacks are
35267 passed in a @code{struct gdb_symbol_callbacks} and for @code{unwind}
35268 and @code{get_frame_id}, in a @code{struct gdb_unwind_callbacks}.
35269 @code{struct gdb_symbol_callbacks} has callbacks to create new object
35270 files and new symbol tables inside those object files. @code{struct
35271 gdb_unwind_callbacks} has callbacks to read registers off the current
35272 frame and to write out the values of the registers in the previous
35273 frame. Both have a callback (@code{target_read}) to read bytes off the
35274 target's address space.
35276 @node In-Process Agent
35277 @chapter In-Process Agent
35278 @cindex debugging agent
35279 The traditional debugging model is conceptually low-speed, but works fine,
35280 because most bugs can be reproduced in debugging-mode execution. However,
35281 as multi-core or many-core processors are becoming mainstream, and
35282 multi-threaded programs become more and more popular, there should be more
35283 and more bugs that only manifest themselves at normal-mode execution, for
35284 example, thread races, because debugger's interference with the program's
35285 timing may conceal the bugs. On the other hand, in some applications,
35286 it is not feasible for the debugger to interrupt the program's execution
35287 long enough for the developer to learn anything helpful about its behavior.
35288 If the program's correctness depends on its real-time behavior, delays
35289 introduced by a debugger might cause the program to fail, even when the
35290 code itself is correct. It is useful to be able to observe the program's
35291 behavior without interrupting it.
35293 Therefore, traditional debugging model is too intrusive to reproduce
35294 some bugs. In order to reduce the interference with the program, we can
35295 reduce the number of operations performed by debugger. The
35296 @dfn{In-Process Agent}, a shared library, is running within the same
35297 process with inferior, and is able to perform some debugging operations
35298 itself. As a result, debugger is only involved when necessary, and
35299 performance of debugging can be improved accordingly. Note that
35300 interference with program can be reduced but can't be removed completely,
35301 because the in-process agent will still stop or slow down the program.
35303 The in-process agent can interpret and execute Agent Expressions
35304 (@pxref{Agent Expressions}) during performing debugging operations. The
35305 agent expressions can be used for different purposes, such as collecting
35306 data in tracepoints, and condition evaluation in breakpoints.
35308 @anchor{Control Agent}
35309 You can control whether the in-process agent is used as an aid for
35310 debugging with the following commands:
35313 @kindex set agent on
35315 Causes the in-process agent to perform some operations on behalf of the
35316 debugger. Just which operations requested by the user will be done
35317 by the in-process agent depends on the its capabilities. For example,
35318 if you request to evaluate breakpoint conditions in the in-process agent,
35319 and the in-process agent has such capability as well, then breakpoint
35320 conditions will be evaluated in the in-process agent.
35322 @kindex set agent off
35323 @item set agent off
35324 Disables execution of debugging operations by the in-process agent. All
35325 of the operations will be performed by @value{GDBN}.
35329 Display the current setting of execution of debugging operations by
35330 the in-process agent.
35334 * In-Process Agent Protocol::
35337 @node In-Process Agent Protocol
35338 @section In-Process Agent Protocol
35339 @cindex in-process agent protocol
35341 The in-process agent is able to communicate with both @value{GDBN} and
35342 GDBserver (@pxref{In-Process Agent}). This section documents the protocol
35343 used for communications between @value{GDBN} or GDBserver and the IPA.
35344 In general, @value{GDBN} or GDBserver sends commands
35345 (@pxref{IPA Protocol Commands}) and data to in-process agent, and then
35346 in-process agent replies back with the return result of the command, or
35347 some other information. The data sent to in-process agent is composed
35348 of primitive data types, such as 4-byte or 8-byte type, and composite
35349 types, which are called objects (@pxref{IPA Protocol Objects}).
35352 * IPA Protocol Objects::
35353 * IPA Protocol Commands::
35356 @node IPA Protocol Objects
35357 @subsection IPA Protocol Objects
35358 @cindex ipa protocol objects
35360 The commands sent to and results received from agent may contain some
35361 complex data types called @dfn{objects}.
35363 The in-process agent is running on the same machine with @value{GDBN}
35364 or GDBserver, so it doesn't have to handle as much differences between
35365 two ends as remote protocol (@pxref{Remote Protocol}) tries to handle.
35366 However, there are still some differences of two ends in two processes:
35370 word size. On some 64-bit machines, @value{GDBN} or GDBserver can be
35371 compiled as a 64-bit executable, while in-process agent is a 32-bit one.
35373 ABI. Some machines may have multiple types of ABI, @value{GDBN} or
35374 GDBserver is compiled with one, and in-process agent is compiled with
35378 Here are the IPA Protocol Objects:
35382 agent expression object. It represents an agent expression
35383 (@pxref{Agent Expressions}).
35384 @anchor{agent expression object}
35386 tracepoint action object. It represents a tracepoint action
35387 (@pxref{Tracepoint Actions,,Tracepoint Action Lists}) to collect registers,
35388 memory, static trace data and to evaluate expression.
35389 @anchor{tracepoint action object}
35391 tracepoint object. It represents a tracepoint (@pxref{Tracepoints}).
35392 @anchor{tracepoint object}
35396 The following table describes important attributes of each IPA protocol
35399 @multitable @columnfractions .30 .20 .50
35400 @headitem Name @tab Size @tab Description
35401 @item @emph{agent expression object} @tab @tab
35402 @item length @tab 4 @tab length of bytes code
35403 @item byte code @tab @var{length} @tab contents of byte code
35404 @item @emph{tracepoint action for collecting memory} @tab @tab
35405 @item 'M' @tab 1 @tab type of tracepoint action
35406 @item addr @tab 8 @tab if @var{basereg} is @samp{-1}, @var{addr} is the
35407 address of the lowest byte to collect, otherwise @var{addr} is the offset
35408 of @var{basereg} for memory collecting.
35409 @item len @tab 8 @tab length of memory for collecting
35410 @item basereg @tab 4 @tab the register number containing the starting
35411 memory address for collecting.
35412 @item @emph{tracepoint action for collecting registers} @tab @tab
35413 @item 'R' @tab 1 @tab type of tracepoint action
35414 @item @emph{tracepoint action for collecting static trace data} @tab @tab
35415 @item 'L' @tab 1 @tab type of tracepoint action
35416 @item @emph{tracepoint action for expression evaluation} @tab @tab
35417 @item 'X' @tab 1 @tab type of tracepoint action
35418 @item agent expression @tab length of @tab @ref{agent expression object}
35419 @item @emph{tracepoint object} @tab @tab
35420 @item number @tab 4 @tab number of tracepoint
35421 @item address @tab 8 @tab address of tracepoint inserted on
35422 @item type @tab 4 @tab type of tracepoint
35423 @item enabled @tab 1 @tab enable or disable of tracepoint
35424 @item step_count @tab 8 @tab step
35425 @item pass_count @tab 8 @tab pass
35426 @item numactions @tab 4 @tab number of tracepoint actions
35427 @item hit count @tab 8 @tab hit count
35428 @item trace frame usage @tab 8 @tab trace frame usage
35429 @item compiled_cond @tab 8 @tab compiled condition
35430 @item orig_size @tab 8 @tab orig size
35431 @item condition @tab 4 if condition is NULL otherwise length of
35432 @ref{agent expression object}
35433 @tab zero if condition is NULL, otherwise is
35434 @ref{agent expression object}
35435 @item actions @tab variable
35436 @tab numactions number of @ref{tracepoint action object}
35439 @node IPA Protocol Commands
35440 @subsection IPA Protocol Commands
35441 @cindex ipa protocol commands
35443 The spaces in each command are delimiters to ease reading this commands
35444 specification. They don't exist in real commands.
35448 @item FastTrace:@var{tracepoint_object} @var{gdb_jump_pad_head}
35449 Installs a new fast tracepoint described by @var{tracepoint_object}
35450 (@pxref{tracepoint object}). The @var{gdb_jump_pad_head}, 8-byte long, is the
35451 head of @dfn{jumppad}, which is used to jump to data collection routine
35456 @item OK @var{target_address} @var{gdb_jump_pad_head} @var{fjump_size} @var{fjump}
35457 @var{target_address} is address of tracepoint in the inferior.
35458 The @var{gdb_jump_pad_head} is updated head of jumppad. Both of
35459 @var{target_address} and @var{gdb_jump_pad_head} are 8-byte long.
35460 The @var{fjump} contains a sequence of instructions jump to jumppad entry.
35461 The @var{fjump_size}, 4-byte long, is the size of @var{fjump}.
35468 Closes the in-process agent. This command is sent when @value{GDBN} or GDBserver
35469 is about to kill inferiors.
35477 @item probe_marker_at:@var{address}
35478 Asks in-process agent to probe the marker at @var{address}.
35485 @item unprobe_marker_at:@var{address}
35486 Asks in-process agent to unprobe the marker at @var{address}.
35490 @chapter Reporting Bugs in @value{GDBN}
35491 @cindex bugs in @value{GDBN}
35492 @cindex reporting bugs in @value{GDBN}
35494 Your bug reports play an essential role in making @value{GDBN} reliable.
35496 Reporting a bug may help you by bringing a solution to your problem, or it
35497 may not. But in any case the principal function of a bug report is to help
35498 the entire community by making the next version of @value{GDBN} work better. Bug
35499 reports are your contribution to the maintenance of @value{GDBN}.
35501 In order for a bug report to serve its purpose, you must include the
35502 information that enables us to fix the bug.
35505 * Bug Criteria:: Have you found a bug?
35506 * Bug Reporting:: How to report bugs
35510 @section Have You Found a Bug?
35511 @cindex bug criteria
35513 If you are not sure whether you have found a bug, here are some guidelines:
35516 @cindex fatal signal
35517 @cindex debugger crash
35518 @cindex crash of debugger
35520 If the debugger gets a fatal signal, for any input whatever, that is a
35521 @value{GDBN} bug. Reliable debuggers never crash.
35523 @cindex error on valid input
35525 If @value{GDBN} produces an error message for valid input, that is a
35526 bug. (Note that if you're cross debugging, the problem may also be
35527 somewhere in the connection to the target.)
35529 @cindex invalid input
35531 If @value{GDBN} does not produce an error message for invalid input,
35532 that is a bug. However, you should note that your idea of
35533 ``invalid input'' might be our idea of ``an extension'' or ``support
35534 for traditional practice''.
35537 If you are an experienced user of debugging tools, your suggestions
35538 for improvement of @value{GDBN} are welcome in any case.
35541 @node Bug Reporting
35542 @section How to Report Bugs
35543 @cindex bug reports
35544 @cindex @value{GDBN} bugs, reporting
35546 A number of companies and individuals offer support for @sc{gnu} products.
35547 If you obtained @value{GDBN} from a support organization, we recommend you
35548 contact that organization first.
35550 You can find contact information for many support companies and
35551 individuals in the file @file{etc/SERVICE} in the @sc{gnu} Emacs
35553 @c should add a web page ref...
35556 @ifset BUGURL_DEFAULT
35557 In any event, we also recommend that you submit bug reports for
35558 @value{GDBN}. The preferred method is to submit them directly using
35559 @uref{http://www.gnu.org/software/gdb/bugs/, @value{GDBN}'s Bugs web
35560 page}. Alternatively, the @email{bug-gdb@@gnu.org, e-mail gateway} can
35563 @strong{Do not send bug reports to @samp{info-gdb}, or to
35564 @samp{help-gdb}, or to any newsgroups.} Most users of @value{GDBN} do
35565 not want to receive bug reports. Those that do have arranged to receive
35568 The mailing list @samp{bug-gdb} has a newsgroup @samp{gnu.gdb.bug} which
35569 serves as a repeater. The mailing list and the newsgroup carry exactly
35570 the same messages. Often people think of posting bug reports to the
35571 newsgroup instead of mailing them. This appears to work, but it has one
35572 problem which can be crucial: a newsgroup posting often lacks a mail
35573 path back to the sender. Thus, if we need to ask for more information,
35574 we may be unable to reach you. For this reason, it is better to send
35575 bug reports to the mailing list.
35577 @ifclear BUGURL_DEFAULT
35578 In any event, we also recommend that you submit bug reports for
35579 @value{GDBN} to @value{BUGURL}.
35583 The fundamental principle of reporting bugs usefully is this:
35584 @strong{report all the facts}. If you are not sure whether to state a
35585 fact or leave it out, state it!
35587 Often people omit facts because they think they know what causes the
35588 problem and assume that some details do not matter. Thus, you might
35589 assume that the name of the variable you use in an example does not matter.
35590 Well, probably it does not, but one cannot be sure. Perhaps the bug is a
35591 stray memory reference which happens to fetch from the location where that
35592 name is stored in memory; perhaps, if the name were different, the contents
35593 of that location would fool the debugger into doing the right thing despite
35594 the bug. Play it safe and give a specific, complete example. That is the
35595 easiest thing for you to do, and the most helpful.
35597 Keep in mind that the purpose of a bug report is to enable us to fix the
35598 bug. It may be that the bug has been reported previously, but neither
35599 you nor we can know that unless your bug report is complete and
35602 Sometimes people give a few sketchy facts and ask, ``Does this ring a
35603 bell?'' Those bug reports are useless, and we urge everyone to
35604 @emph{refuse to respond to them} except to chide the sender to report
35607 To enable us to fix the bug, you should include all these things:
35611 The version of @value{GDBN}. @value{GDBN} announces it if you start
35612 with no arguments; you can also print it at any time using @code{show
35615 Without this, we will not know whether there is any point in looking for
35616 the bug in the current version of @value{GDBN}.
35619 The type of machine you are using, and the operating system name and
35623 The details of the @value{GDBN} build-time configuration.
35624 @value{GDBN} shows these details if you invoke it with the
35625 @option{--configuration} command-line option, or if you type
35626 @code{show configuration} at @value{GDBN}'s prompt.
35629 What compiler (and its version) was used to compile @value{GDBN}---e.g.@:
35630 ``@value{GCC}--2.8.1''.
35633 What compiler (and its version) was used to compile the program you are
35634 debugging---e.g.@: ``@value{GCC}--2.8.1'', or ``HP92453-01 A.10.32.03 HP
35635 C Compiler''. For @value{NGCC}, you can say @kbd{@value{GCC} --version}
35636 to get this information; for other compilers, see the documentation for
35640 The command arguments you gave the compiler to compile your example and
35641 observe the bug. For example, did you use @samp{-O}? To guarantee
35642 you will not omit something important, list them all. A copy of the
35643 Makefile (or the output from make) is sufficient.
35645 If we were to try to guess the arguments, we would probably guess wrong
35646 and then we might not encounter the bug.
35649 A complete input script, and all necessary source files, that will
35653 A description of what behavior you observe that you believe is
35654 incorrect. For example, ``It gets a fatal signal.''
35656 Of course, if the bug is that @value{GDBN} gets a fatal signal, then we
35657 will certainly notice it. But if the bug is incorrect output, we might
35658 not notice unless it is glaringly wrong. You might as well not give us
35659 a chance to make a mistake.
35661 Even if the problem you experience is a fatal signal, you should still
35662 say so explicitly. Suppose something strange is going on, such as, your
35663 copy of @value{GDBN} is out of synch, or you have encountered a bug in
35664 the C library on your system. (This has happened!) Your copy might
35665 crash and ours would not. If you told us to expect a crash, then when
35666 ours fails to crash, we would know that the bug was not happening for
35667 us. If you had not told us to expect a crash, then we would not be able
35668 to draw any conclusion from our observations.
35671 @cindex recording a session script
35672 To collect all this information, you can use a session recording program
35673 such as @command{script}, which is available on many Unix systems.
35674 Just run your @value{GDBN} session inside @command{script} and then
35675 include the @file{typescript} file with your bug report.
35677 Another way to record a @value{GDBN} session is to run @value{GDBN}
35678 inside Emacs and then save the entire buffer to a file.
35681 If you wish to suggest changes to the @value{GDBN} source, send us context
35682 diffs. If you even discuss something in the @value{GDBN} source, refer to
35683 it by context, not by line number.
35685 The line numbers in our development sources will not match those in your
35686 sources. Your line numbers would convey no useful information to us.
35690 Here are some things that are not necessary:
35694 A description of the envelope of the bug.
35696 Often people who encounter a bug spend a lot of time investigating
35697 which changes to the input file will make the bug go away and which
35698 changes will not affect it.
35700 This is often time consuming and not very useful, because the way we
35701 will find the bug is by running a single example under the debugger
35702 with breakpoints, not by pure deduction from a series of examples.
35703 We recommend that you save your time for something else.
35705 Of course, if you can find a simpler example to report @emph{instead}
35706 of the original one, that is a convenience for us. Errors in the
35707 output will be easier to spot, running under the debugger will take
35708 less time, and so on.
35710 However, simplification is not vital; if you do not want to do this,
35711 report the bug anyway and send us the entire test case you used.
35714 A patch for the bug.
35716 A patch for the bug does help us if it is a good one. But do not omit
35717 the necessary information, such as the test case, on the assumption that
35718 a patch is all we need. We might see problems with your patch and decide
35719 to fix the problem another way, or we might not understand it at all.
35721 Sometimes with a program as complicated as @value{GDBN} it is very hard to
35722 construct an example that will make the program follow a certain path
35723 through the code. If you do not send us the example, we will not be able
35724 to construct one, so we will not be able to verify that the bug is fixed.
35726 And if we cannot understand what bug you are trying to fix, or why your
35727 patch should be an improvement, we will not install it. A test case will
35728 help us to understand.
35731 A guess about what the bug is or what it depends on.
35733 Such guesses are usually wrong. Even we cannot guess right about such
35734 things without first using the debugger to find the facts.
35737 @c The readline documentation is distributed with the readline code
35738 @c and consists of the two following files:
35741 @c Use -I with makeinfo to point to the appropriate directory,
35742 @c environment var TEXINPUTS with TeX.
35743 @ifclear SYSTEM_READLINE
35744 @include rluser.texi
35745 @include hsuser.texi
35749 @appendix In Memoriam
35751 The @value{GDBN} project mourns the loss of the following long-time
35756 Fred was a long-standing contributor to @value{GDBN} (1991-2006), and
35757 to Free Software in general. Outside of @value{GDBN}, he was known in
35758 the Amiga world for his series of Fish Disks, and the GeekGadget project.
35760 @item Michael Snyder
35761 Michael was one of the Global Maintainers of the @value{GDBN} project,
35762 with contributions recorded as early as 1996, until 2011. In addition
35763 to his day to day participation, he was a large driving force behind
35764 adding Reverse Debugging to @value{GDBN}.
35767 Beyond their technical contributions to the project, they were also
35768 enjoyable members of the Free Software Community. We will miss them.
35770 @node Formatting Documentation
35771 @appendix Formatting Documentation
35773 @cindex @value{GDBN} reference card
35774 @cindex reference card
35775 The @value{GDBN} 4 release includes an already-formatted reference card, ready
35776 for printing with PostScript or Ghostscript, in the @file{gdb}
35777 subdirectory of the main source directory@footnote{In
35778 @file{gdb-@value{GDBVN}/gdb/refcard.ps} of the version @value{GDBVN}
35779 release.}. If you can use PostScript or Ghostscript with your printer,
35780 you can print the reference card immediately with @file{refcard.ps}.
35782 The release also includes the source for the reference card. You
35783 can format it, using @TeX{}, by typing:
35789 The @value{GDBN} reference card is designed to print in @dfn{landscape}
35790 mode on US ``letter'' size paper;
35791 that is, on a sheet 11 inches wide by 8.5 inches
35792 high. You will need to specify this form of printing as an option to
35793 your @sc{dvi} output program.
35795 @cindex documentation
35797 All the documentation for @value{GDBN} comes as part of the machine-readable
35798 distribution. The documentation is written in Texinfo format, which is
35799 a documentation system that uses a single source file to produce both
35800 on-line information and a printed manual. You can use one of the Info
35801 formatting commands to create the on-line version of the documentation
35802 and @TeX{} (or @code{texi2roff}) to typeset the printed version.
35804 @value{GDBN} includes an already formatted copy of the on-line Info
35805 version of this manual in the @file{gdb} subdirectory. The main Info
35806 file is @file{gdb-@value{GDBVN}/gdb/gdb.info}, and it refers to
35807 subordinate files matching @samp{gdb.info*} in the same directory. If
35808 necessary, you can print out these files, or read them with any editor;
35809 but they are easier to read using the @code{info} subsystem in @sc{gnu}
35810 Emacs or the standalone @code{info} program, available as part of the
35811 @sc{gnu} Texinfo distribution.
35813 If you want to format these Info files yourself, you need one of the
35814 Info formatting programs, such as @code{texinfo-format-buffer} or
35817 If you have @code{makeinfo} installed, and are in the top level
35818 @value{GDBN} source directory (@file{gdb-@value{GDBVN}}, in the case of
35819 version @value{GDBVN}), you can make the Info file by typing:
35826 If you want to typeset and print copies of this manual, you need @TeX{},
35827 a program to print its @sc{dvi} output files, and @file{texinfo.tex}, the
35828 Texinfo definitions file.
35830 @TeX{} is a typesetting program; it does not print files directly, but
35831 produces output files called @sc{dvi} files. To print a typeset
35832 document, you need a program to print @sc{dvi} files. If your system
35833 has @TeX{} installed, chances are it has such a program. The precise
35834 command to use depends on your system; @kbd{lpr -d} is common; another
35835 (for PostScript devices) is @kbd{dvips}. The @sc{dvi} print command may
35836 require a file name without any extension or a @samp{.dvi} extension.
35838 @TeX{} also requires a macro definitions file called
35839 @file{texinfo.tex}. This file tells @TeX{} how to typeset a document
35840 written in Texinfo format. On its own, @TeX{} cannot either read or
35841 typeset a Texinfo file. @file{texinfo.tex} is distributed with GDB
35842 and is located in the @file{gdb-@var{version-number}/texinfo}
35845 If you have @TeX{} and a @sc{dvi} printer program installed, you can
35846 typeset and print this manual. First switch to the @file{gdb}
35847 subdirectory of the main source directory (for example, to
35848 @file{gdb-@value{GDBVN}/gdb}) and type:
35854 Then give @file{gdb.dvi} to your @sc{dvi} printing program.
35856 @node Installing GDB
35857 @appendix Installing @value{GDBN}
35858 @cindex installation
35861 * Requirements:: Requirements for building @value{GDBN}
35862 * Running Configure:: Invoking the @value{GDBN} @file{configure} script
35863 * Separate Objdir:: Compiling @value{GDBN} in another directory
35864 * Config Names:: Specifying names for hosts and targets
35865 * Configure Options:: Summary of options for configure
35866 * System-wide configuration:: Having a system-wide init file
35870 @section Requirements for Building @value{GDBN}
35871 @cindex building @value{GDBN}, requirements for
35873 Building @value{GDBN} requires various tools and packages to be available.
35874 Other packages will be used only if they are found.
35876 @heading Tools/Packages Necessary for Building @value{GDBN}
35878 @item C@t{++}11 compiler
35879 @value{GDBN} is written in C@t{++}11. It should be buildable with any
35880 recent C@t{++}11 compiler, e.g.@: GCC.
35883 @value{GDBN}'s build system relies on features only found in the GNU
35884 make program. Other variants of @code{make} will not work.
35887 @heading Tools/Packages Optional for Building @value{GDBN}
35891 @value{GDBN} can use the Expat XML parsing library. This library may be
35892 included with your operating system distribution; if it is not, you
35893 can get the latest version from @url{http://expat.sourceforge.net}.
35894 The @file{configure} script will search for this library in several
35895 standard locations; if it is installed in an unusual path, you can
35896 use the @option{--with-libexpat-prefix} option to specify its location.
35902 Remote protocol memory maps (@pxref{Memory Map Format})
35904 Target descriptions (@pxref{Target Descriptions})
35906 Remote shared library lists (@xref{Library List Format},
35907 or alternatively @pxref{Library List Format for SVR4 Targets})
35909 MS-Windows shared libraries (@pxref{Shared Libraries})
35911 Traceframe info (@pxref{Traceframe Info Format})
35913 Branch trace (@pxref{Branch Trace Format},
35914 @pxref{Branch Trace Configuration Format})
35918 @value{GDBN} can be scripted using GNU Guile. @xref{Guile}. By
35919 default, @value{GDBN} will be compiled if the Guile libraries are
35920 installed and are found by @file{configure}. You can use the
35921 @code{--with-guile} option to request Guile, and pass either the Guile
35922 version number or the file name of the relevant @code{pkg-config}
35923 program to choose a particular version of Guile.
35926 @value{GDBN}'s features related to character sets (@pxref{Character
35927 Sets}) require a functioning @code{iconv} implementation. If you are
35928 on a GNU system, then this is provided by the GNU C Library. Some
35929 other systems also provide a working @code{iconv}.
35931 If @value{GDBN} is using the @code{iconv} program which is installed
35932 in a non-standard place, you will need to tell @value{GDBN} where to
35933 find it. This is done with @option{--with-iconv-bin} which specifies
35934 the directory that contains the @code{iconv} program. This program is
35935 run in order to make a list of the available character sets.
35937 On systems without @code{iconv}, you can install GNU Libiconv. If
35938 Libiconv is installed in a standard place, @value{GDBN} will
35939 automatically use it if it is needed. If you have previously
35940 installed Libiconv in a non-standard place, you can use the
35941 @option{--with-libiconv-prefix} option to @file{configure}.
35943 @value{GDBN}'s top-level @file{configure} and @file{Makefile} will
35944 arrange to build Libiconv if a directory named @file{libiconv} appears
35945 in the top-most source directory. If Libiconv is built this way, and
35946 if the operating system does not provide a suitable @code{iconv}
35947 implementation, then the just-built library will automatically be used
35948 by @value{GDBN}. One easy way to set this up is to download GNU
35949 Libiconv, unpack it inside the top-level directory of the @value{GDBN}
35950 source tree, and then rename the directory holding the Libiconv source
35951 code to @samp{libiconv}.
35954 @value{GDBN} can support debugging sections that are compressed with
35955 the LZMA library. @xref{MiniDebugInfo}. If this library is not
35956 included with your operating system, you can find it in the xz package
35957 at @url{http://tukaani.org/xz/}. If the LZMA library is available in
35958 the usual place, then the @file{configure} script will use it
35959 automatically. If it is installed in an unusual path, you can use the
35960 @option{--with-lzma-prefix} option to specify its location.
35964 @value{GDBN} can use the GNU MPFR multiple-precision floating-point
35965 library. This library may be included with your operating system
35966 distribution; if it is not, you can get the latest version from
35967 @url{http://www.mpfr.org}. The @file{configure} script will search
35968 for this library in several standard locations; if it is installed
35969 in an unusual path, you can use the @option{--with-libmpfr-prefix}
35970 option to specify its location.
35972 GNU MPFR is used to emulate target floating-point arithmetic during
35973 expression evaluation when the target uses different floating-point
35974 formats than the host. If GNU MPFR it is not available, @value{GDBN}
35975 will fall back to using host floating-point arithmetic.
35978 @value{GDBN} can be scripted using Python language. @xref{Python}.
35979 By default, @value{GDBN} will be compiled if the Python libraries are
35980 installed and are found by @file{configure}. You can use the
35981 @code{--with-python} option to request Python, and pass either the
35982 file name of the relevant @code{python} executable, or the name of the
35983 directory in which Python is installed, to choose a particular
35984 installation of Python.
35987 @cindex compressed debug sections
35988 @value{GDBN} will use the @samp{zlib} library, if available, to read
35989 compressed debug sections. Some linkers, such as GNU gold, are capable
35990 of producing binaries with compressed debug sections. If @value{GDBN}
35991 is compiled with @samp{zlib}, it will be able to read the debug
35992 information in such binaries.
35994 The @samp{zlib} library is likely included with your operating system
35995 distribution; if it is not, you can get the latest version from
35996 @url{http://zlib.net}.
35999 @node Running Configure
36000 @section Invoking the @value{GDBN} @file{configure} Script
36001 @cindex configuring @value{GDBN}
36002 @value{GDBN} comes with a @file{configure} script that automates the process
36003 of preparing @value{GDBN} for installation; you can then use @code{make} to
36004 build the @code{gdb} program.
36006 @c irrelevant in info file; it's as current as the code it lives with.
36007 @footnote{If you have a more recent version of @value{GDBN} than @value{GDBVN},
36008 look at the @file{README} file in the sources; we may have improved the
36009 installation procedures since publishing this manual.}
36012 The @value{GDBN} distribution includes all the source code you need for
36013 @value{GDBN} in a single directory, whose name is usually composed by
36014 appending the version number to @samp{gdb}.
36016 For example, the @value{GDBN} version @value{GDBVN} distribution is in the
36017 @file{gdb-@value{GDBVN}} directory. That directory contains:
36020 @item gdb-@value{GDBVN}/configure @r{(and supporting files)}
36021 script for configuring @value{GDBN} and all its supporting libraries
36023 @item gdb-@value{GDBVN}/gdb
36024 the source specific to @value{GDBN} itself
36026 @item gdb-@value{GDBVN}/bfd
36027 source for the Binary File Descriptor library
36029 @item gdb-@value{GDBVN}/include
36030 @sc{gnu} include files
36032 @item gdb-@value{GDBVN}/libiberty
36033 source for the @samp{-liberty} free software library
36035 @item gdb-@value{GDBVN}/opcodes
36036 source for the library of opcode tables and disassemblers
36038 @item gdb-@value{GDBVN}/readline
36039 source for the @sc{gnu} command-line interface
36042 There may be other subdirectories as well.
36044 The simplest way to configure and build @value{GDBN} is to run @file{configure}
36045 from the @file{gdb-@var{version-number}} source directory, which in
36046 this example is the @file{gdb-@value{GDBVN}} directory.
36048 First switch to the @file{gdb-@var{version-number}} source directory
36049 if you are not already in it; then run @file{configure}. Pass the
36050 identifier for the platform on which @value{GDBN} will run as an
36056 cd gdb-@value{GDBVN}
36061 Running @samp{configure} and then running @code{make} builds the
36062 included supporting libraries, then @code{gdb} itself. The configured
36063 source files, and the binaries, are left in the corresponding source
36067 @file{configure} is a Bourne-shell (@code{/bin/sh}) script; if your
36068 system does not recognize this automatically when you run a different
36069 shell, you may need to run @code{sh} on it explicitly:
36075 You should run the @file{configure} script from the top directory in the
36076 source tree, the @file{gdb-@var{version-number}} directory. If you run
36077 @file{configure} from one of the subdirectories, you will configure only
36078 that subdirectory. That is usually not what you want. In particular,
36079 if you run the first @file{configure} from the @file{gdb} subdirectory
36080 of the @file{gdb-@var{version-number}} directory, you will omit the
36081 configuration of @file{bfd}, @file{readline}, and other sibling
36082 directories of the @file{gdb} subdirectory. This leads to build errors
36083 about missing include files such as @file{bfd/bfd.h}.
36085 You can install @code{@value{GDBN}} anywhere. The best way to do this
36086 is to pass the @code{--prefix} option to @code{configure}, and then
36087 install it with @code{make install}.
36089 @node Separate Objdir
36090 @section Compiling @value{GDBN} in Another Directory
36092 If you want to run @value{GDBN} versions for several host or target machines,
36093 you need a different @code{gdb} compiled for each combination of
36094 host and target. @file{configure} is designed to make this easy by
36095 allowing you to generate each configuration in a separate subdirectory,
36096 rather than in the source directory. If your @code{make} program
36097 handles the @samp{VPATH} feature (@sc{gnu} @code{make} does), running
36098 @code{make} in each of these directories builds the @code{gdb}
36099 program specified there.
36101 To build @code{gdb} in a separate directory, run @file{configure}
36102 with the @samp{--srcdir} option to specify where to find the source.
36103 (You also need to specify a path to find @file{configure}
36104 itself from your working directory. If the path to @file{configure}
36105 would be the same as the argument to @samp{--srcdir}, you can leave out
36106 the @samp{--srcdir} option; it is assumed.)
36108 For example, with version @value{GDBVN}, you can build @value{GDBN} in a
36109 separate directory for a Sun 4 like this:
36113 cd gdb-@value{GDBVN}
36116 ../gdb-@value{GDBVN}/configure
36121 When @file{configure} builds a configuration using a remote source
36122 directory, it creates a tree for the binaries with the same structure
36123 (and using the same names) as the tree under the source directory. In
36124 the example, you'd find the Sun 4 library @file{libiberty.a} in the
36125 directory @file{gdb-sun4/libiberty}, and @value{GDBN} itself in
36126 @file{gdb-sun4/gdb}.
36128 Make sure that your path to the @file{configure} script has just one
36129 instance of @file{gdb} in it. If your path to @file{configure} looks
36130 like @file{../gdb-@value{GDBVN}/gdb/configure}, you are configuring only
36131 one subdirectory of @value{GDBN}, not the whole package. This leads to
36132 build errors about missing include files such as @file{bfd/bfd.h}.
36134 One popular reason to build several @value{GDBN} configurations in separate
36135 directories is to configure @value{GDBN} for cross-compiling (where
36136 @value{GDBN} runs on one machine---the @dfn{host}---while debugging
36137 programs that run on another machine---the @dfn{target}).
36138 You specify a cross-debugging target by
36139 giving the @samp{--target=@var{target}} option to @file{configure}.
36141 When you run @code{make} to build a program or library, you must run
36142 it in a configured directory---whatever directory you were in when you
36143 called @file{configure} (or one of its subdirectories).
36145 The @code{Makefile} that @file{configure} generates in each source
36146 directory also runs recursively. If you type @code{make} in a source
36147 directory such as @file{gdb-@value{GDBVN}} (or in a separate configured
36148 directory configured with @samp{--srcdir=@var{dirname}/gdb-@value{GDBVN}}), you
36149 will build all the required libraries, and then build GDB.
36151 When you have multiple hosts or targets configured in separate
36152 directories, you can run @code{make} on them in parallel (for example,
36153 if they are NFS-mounted on each of the hosts); they will not interfere
36157 @section Specifying Names for Hosts and Targets
36159 The specifications used for hosts and targets in the @file{configure}
36160 script are based on a three-part naming scheme, but some short predefined
36161 aliases are also supported. The full naming scheme encodes three pieces
36162 of information in the following pattern:
36165 @var{architecture}-@var{vendor}-@var{os}
36168 For example, you can use the alias @code{sun4} as a @var{host} argument,
36169 or as the value for @var{target} in a @code{--target=@var{target}}
36170 option. The equivalent full name is @samp{sparc-sun-sunos4}.
36172 The @file{configure} script accompanying @value{GDBN} does not provide
36173 any query facility to list all supported host and target names or
36174 aliases. @file{configure} calls the Bourne shell script
36175 @code{config.sub} to map abbreviations to full names; you can read the
36176 script, if you wish, or you can use it to test your guesses on
36177 abbreviations---for example:
36180 % sh config.sub i386-linux
36182 % sh config.sub alpha-linux
36183 alpha-unknown-linux-gnu
36184 % sh config.sub hp9k700
36186 % sh config.sub sun4
36187 sparc-sun-sunos4.1.1
36188 % sh config.sub sun3
36189 m68k-sun-sunos4.1.1
36190 % sh config.sub i986v
36191 Invalid configuration `i986v': machine `i986v' not recognized
36195 @code{config.sub} is also distributed in the @value{GDBN} source
36196 directory (@file{gdb-@value{GDBVN}}, for version @value{GDBVN}).
36198 @node Configure Options
36199 @section @file{configure} Options
36201 Here is a summary of the @file{configure} options and arguments that
36202 are most often useful for building @value{GDBN}. @file{configure}
36203 also has several other options not listed here. @inforef{Running
36204 configure scripts,,autoconf.info}, for a full
36205 explanation of @file{configure}.
36208 configure @r{[}--help@r{]}
36209 @r{[}--prefix=@var{dir}@r{]}
36210 @r{[}--exec-prefix=@var{dir}@r{]}
36211 @r{[}--srcdir=@var{dirname}@r{]}
36212 @r{[}--target=@var{target}@r{]}
36216 You may introduce options with a single @samp{-} rather than
36217 @samp{--} if you prefer; but you may abbreviate option names if you use
36222 Display a quick summary of how to invoke @file{configure}.
36224 @item --prefix=@var{dir}
36225 Configure the source to install programs and files under directory
36228 @item --exec-prefix=@var{dir}
36229 Configure the source to install programs under directory
36232 @c avoid splitting the warning from the explanation:
36234 @item --srcdir=@var{dirname}
36235 Use this option to make configurations in directories separate from the
36236 @value{GDBN} source directories. Among other things, you can use this to
36237 build (or maintain) several configurations simultaneously, in separate
36238 directories. @file{configure} writes configuration-specific files in
36239 the current directory, but arranges for them to use the source in the
36240 directory @var{dirname}. @file{configure} creates directories under
36241 the working directory in parallel to the source directories below
36244 @item --target=@var{target}
36245 Configure @value{GDBN} for cross-debugging programs running on the specified
36246 @var{target}. Without this option, @value{GDBN} is configured to debug
36247 programs that run on the same machine (@var{host}) as @value{GDBN} itself.
36249 There is no convenient way to generate a list of all available
36250 targets. Also see the @code{--enable-targets} option, below.
36253 There are many other options that are specific to @value{GDBN}. This
36254 lists just the most common ones; there are some very specialized
36255 options not described here.
36258 @item --enable-targets=@r{[}@var{target}@r{]}@dots{}
36259 @itemx --enable-targets=all
36260 Configure @value{GDBN} for cross-debugging programs running on the
36261 specified list of targets. The special value @samp{all} configures
36262 @value{GDBN} for debugging programs running on any target it supports.
36264 @item --with-gdb-datadir=@var{path}
36265 Set the @value{GDBN}-specific data directory. @value{GDBN} will look
36266 here for certain supporting files or scripts. This defaults to the
36267 @file{gdb} subdirectory of @samp{datadi} (which can be set using
36270 @item --with-relocated-sources=@var{dir}
36271 Sets up the default source path substitution rule so that directory
36272 names recorded in debug information will be automatically adjusted for
36273 any directory under @var{dir}. @var{dir} should be a subdirectory of
36274 @value{GDBN}'s configured prefix, the one mentioned in the
36275 @code{--prefix} or @code{--exec-prefix} options to configure. This
36276 option is useful if GDB is supposed to be moved to a different place
36279 @item --enable-64-bit-bfd
36280 Enable 64-bit support in BFD on 32-bit hosts.
36282 @item --disable-gdbmi
36283 Build @value{GDBN} without the GDB/MI machine interface
36287 Build @value{GDBN} with the text-mode full-screen user interface
36288 (TUI). Requires a curses library (ncurses and cursesX are also
36291 @item --with-curses
36292 Use the curses library instead of the termcap library, for text-mode
36293 terminal operations.
36295 @item --with-libunwind-ia64
36296 Use the libunwind library for unwinding function call stack on ia64
36297 target platforms. See http://www.nongnu.org/libunwind/index.html for
36300 @item --with-system-readline
36301 Use the readline library installed on the host, rather than the
36302 library supplied as part of @value{GDBN}.
36304 @item --with-system-zlib
36305 Use the zlib library installed on the host, rather than the library
36306 supplied as part of @value{GDBN}.
36309 Build @value{GDBN} with Expat, a library for XML parsing. (Done by
36310 default if libexpat is installed and found at configure time.) This
36311 library is used to read XML files supplied with @value{GDBN}. If it
36312 is unavailable, some features, such as remote protocol memory maps,
36313 target descriptions, and shared library lists, that are based on XML
36314 files, will not be available in @value{GDBN}. If your host does not
36315 have libexpat installed, you can get the latest version from
36316 `http://expat.sourceforge.net'.
36318 @item --with-libiconv-prefix@r{[}=@var{dir}@r{]}
36320 Build @value{GDBN} with GNU libiconv, a character set encoding
36321 conversion library. This is not done by default, as on GNU systems
36322 the @code{iconv} that is built in to the C library is sufficient. If
36323 your host does not have a working @code{iconv}, you can get the latest
36324 version of GNU iconv from `https://www.gnu.org/software/libiconv/'.
36326 @value{GDBN}'s build system also supports building GNU libiconv as
36327 part of the overall build. @xref{Requirements}.
36330 Build @value{GDBN} with LZMA, a compression library. (Done by default
36331 if liblzma is installed and found at configure time.) LZMA is used by
36332 @value{GDBN}'s "mini debuginfo" feature, which is only useful on
36333 platforms using the ELF object file format. If your host does not
36334 have liblzma installed, you can get the latest version from
36335 `https://tukaani.org/xz/'.
36338 Build @value{GDBN} with GNU MPFR, a library for multiple-precision
36339 floating-point computation with correct rounding. (Done by default if
36340 GNU MPFR is installed and found at configure time.) This library is
36341 used to emulate target floating-point arithmetic during expression
36342 evaluation when the target uses different floating-point formats than
36343 the host. If GNU MPFR is not available, @value{GDBN} will fall back
36344 to using host floating-point arithmetic. If your host does not have
36345 GNU MPFR installed, you can get the latest version from
36346 `http://www.mpfr.org'.
36348 @item --with-python@r{[}=@var{python}@r{]}
36349 Build @value{GDBN} with Python scripting support. (Done by default if
36350 libpython is present and found at configure time.) Python makes
36351 @value{GDBN} scripting much more powerful than the restricted CLI
36352 scripting language. If your host does not have Python installed, you
36353 can find it on `http://www.python.org/download/'. The oldest version
36354 of Python supported by GDB is 2.6. The optional argument @var{python}
36355 is used to find the Python headers and libraries. It can be either
36356 the name of a Python executable, or the name of the directory in which
36357 Python is installed.
36359 @item --with-guile[=GUILE]'
36360 Build @value{GDBN} with GNU Guile scripting support. (Done by default
36361 if libguile is present and found at configure time.) If your host
36362 does not have Guile installed, you can find it at
36363 `https://www.gnu.org/software/guile/'. The optional argument GUILE
36364 can be a version number, which will cause @code{configure} to try to
36365 use that version of Guile; or the file name of a @code{pkg-config}
36366 executable, which will be queried to find the information needed to
36367 compile and link against Guile.
36369 @item --without-included-regex
36370 Don't use the regex library included with @value{GDBN} (as part of the
36371 libiberty library). This is the default on hosts with version 2 of
36374 @item --with-sysroot=@var{dir}
36375 Use @var{dir} as the default system root directory for libraries whose
36376 file names begin with @file{/lib}' or @file{/usr/lib'}. (The value of
36377 @var{dir} can be modified at run time by using the @command{set
36378 sysroot} command.) If @var{dir} is under the @value{GDBN} configured
36379 prefix (set with @code{--prefix} or @code{--exec-prefix options}, the
36380 default system root will be automatically adjusted if and when
36381 @value{GDBN} is moved to a different location.
36383 @item --with-system-gdbinit=@var{file}
36384 Configure @value{GDBN} to automatically load a system-wide init file.
36385 @var{file} should be an absolute file name. If @var{file} is in a
36386 directory under the configured prefix, and @value{GDBN} is moved to
36387 another location after being built, the location of the system-wide
36388 init file will be adjusted accordingly.
36390 @item --enable-build-warnings
36391 When building the @value{GDBN} sources, ask the compiler to warn about
36392 any code which looks even vaguely suspicious. It passes many
36393 different warning flags, depending on the exact version of the
36394 compiler you are using.
36396 @item --enable-werror
36397 Treat compiler warnings as werrors. It adds the @code{-Werror} flag
36398 to the compiler, which will fail the compilation if the compiler
36399 outputs any warning messages.
36401 @item --enable-ubsan
36402 Enable the GCC undefined behavior sanitizer. This is disabled by
36403 default, but passing @code{--enable-ubsan=yes} or
36404 @code{--enable-ubsan=auto} to @code{configure} will enable it. The
36405 undefined behavior sanitizer checks for C@t{++} undefined behavior.
36406 It has a performance cost, so if you are looking at @value{GDBN}'s
36407 performance, you should disable it. The undefined behavior sanitizer
36408 was first introduced in GCC 4.9.
36411 @node System-wide configuration
36412 @section System-wide configuration and settings
36413 @cindex system-wide init file
36415 @value{GDBN} can be configured to have a system-wide init file;
36416 this file will be read and executed at startup (@pxref{Startup, , What
36417 @value{GDBN} does during startup}).
36419 Here is the corresponding configure option:
36422 @item --with-system-gdbinit=@var{file}
36423 Specify that the default location of the system-wide init file is
36427 If @value{GDBN} has been configured with the option @option{--prefix=$prefix},
36428 it may be subject to relocation. Two possible cases:
36432 If the default location of this init file contains @file{$prefix},
36433 it will be subject to relocation. Suppose that the configure options
36434 are @option{--prefix=$prefix --with-system-gdbinit=$prefix/etc/gdbinit};
36435 if @value{GDBN} is moved from @file{$prefix} to @file{$install}, the system
36436 init file is looked for as @file{$install/etc/gdbinit} instead of
36437 @file{$prefix/etc/gdbinit}.
36440 By contrast, if the default location does not contain the prefix,
36441 it will not be relocated. E.g.@: if @value{GDBN} has been configured with
36442 @option{--prefix=/usr/local --with-system-gdbinit=/usr/share/gdb/gdbinit},
36443 then @value{GDBN} will always look for @file{/usr/share/gdb/gdbinit},
36444 wherever @value{GDBN} is installed.
36447 If the configured location of the system-wide init file (as given by the
36448 @option{--with-system-gdbinit} option at configure time) is in the
36449 data-directory (as specified by @option{--with-gdb-datadir} at configure
36450 time) or in one of its subdirectories, then @value{GDBN} will look for the
36451 system-wide init file in the directory specified by the
36452 @option{--data-directory} command-line option.
36453 Note that the system-wide init file is only read once, during @value{GDBN}
36454 initialization. If the data-directory is changed after @value{GDBN} has
36455 started with the @code{set data-directory} command, the file will not be
36459 * System-wide Configuration Scripts:: Installed System-wide Configuration Scripts
36462 @node System-wide Configuration Scripts
36463 @subsection Installed System-wide Configuration Scripts
36464 @cindex system-wide configuration scripts
36466 The @file{system-gdbinit} directory, located inside the data-directory
36467 (as specified by @option{--with-gdb-datadir} at configure time) contains
36468 a number of scripts which can be used as system-wide init files. To
36469 automatically source those scripts at startup, @value{GDBN} should be
36470 configured with @option{--with-system-gdbinit}. Otherwise, any user
36471 should be able to source them by hand as needed.
36473 The following scripts are currently available:
36476 @item @file{elinos.py}
36478 @cindex ELinOS system-wide configuration script
36479 This script is useful when debugging a program on an ELinOS target.
36480 It takes advantage of the environment variables defined in a standard
36481 ELinOS environment in order to determine the location of the system
36482 shared libraries, and then sets the @samp{solib-absolute-prefix}
36483 and @samp{solib-search-path} variables appropriately.
36485 @item @file{wrs-linux.py}
36486 @pindex wrs-linux.py
36487 @cindex Wind River Linux system-wide configuration script
36488 This script is useful when debugging a program on a target running
36489 Wind River Linux. It expects the @env{ENV_PREFIX} to be set to
36490 the host-side sysroot used by the target system.
36494 @node Maintenance Commands
36495 @appendix Maintenance Commands
36496 @cindex maintenance commands
36497 @cindex internal commands
36499 In addition to commands intended for @value{GDBN} users, @value{GDBN}
36500 includes a number of commands intended for @value{GDBN} developers,
36501 that are not documented elsewhere in this manual. These commands are
36502 provided here for reference. (For commands that turn on debugging
36503 messages, see @ref{Debugging Output}.)
36506 @kindex maint agent
36507 @kindex maint agent-eval
36508 @item maint agent @r{[}-at @var{location}@r{,}@r{]} @var{expression}
36509 @itemx maint agent-eval @r{[}-at @var{location}@r{,}@r{]} @var{expression}
36510 Translate the given @var{expression} into remote agent bytecodes.
36511 This command is useful for debugging the Agent Expression mechanism
36512 (@pxref{Agent Expressions}). The @samp{agent} version produces an
36513 expression useful for data collection, such as by tracepoints, while
36514 @samp{maint agent-eval} produces an expression that evaluates directly
36515 to a result. For instance, a collection expression for @code{globa +
36516 globb} will include bytecodes to record four bytes of memory at each
36517 of the addresses of @code{globa} and @code{globb}, while discarding
36518 the result of the addition, while an evaluation expression will do the
36519 addition and return the sum.
36520 If @code{-at} is given, generate remote agent bytecode for @var{location}.
36521 If not, generate remote agent bytecode for current frame PC address.
36523 @kindex maint agent-printf
36524 @item maint agent-printf @var{format},@var{expr},...
36525 Translate the given format string and list of argument expressions
36526 into remote agent bytecodes and display them as a disassembled list.
36527 This command is useful for debugging the agent version of dynamic
36528 printf (@pxref{Dynamic Printf}).
36530 @kindex maint info breakpoints
36531 @item @anchor{maint info breakpoints}maint info breakpoints
36532 Using the same format as @samp{info breakpoints}, display both the
36533 breakpoints you've set explicitly, and those @value{GDBN} is using for
36534 internal purposes. Internal breakpoints are shown with negative
36535 breakpoint numbers. The type column identifies what kind of breakpoint
36540 Normal, explicitly set breakpoint.
36543 Normal, explicitly set watchpoint.
36546 Internal breakpoint, used to handle correctly stepping through
36547 @code{longjmp} calls.
36549 @item longjmp resume
36550 Internal breakpoint at the target of a @code{longjmp}.
36553 Temporary internal breakpoint used by the @value{GDBN} @code{until} command.
36556 Temporary internal breakpoint used by the @value{GDBN} @code{finish} command.
36559 Shared library events.
36563 @kindex maint info btrace
36564 @item maint info btrace
36565 Pint information about raw branch tracing data.
36567 @kindex maint btrace packet-history
36568 @item maint btrace packet-history
36569 Print the raw branch trace packets that are used to compute the
36570 execution history for the @samp{record btrace} command. Both the
36571 information and the format in which it is printed depend on the btrace
36576 For the BTS recording format, print a list of blocks of sequential
36577 code. For each block, the following information is printed:
36581 Newer blocks have higher numbers. The oldest block has number zero.
36582 @item Lowest @samp{PC}
36583 @item Highest @samp{PC}
36587 For the Intel Processor Trace recording format, print a list of
36588 Intel Processor Trace packets. For each packet, the following
36589 information is printed:
36592 @item Packet number
36593 Newer packets have higher numbers. The oldest packet has number zero.
36595 The packet's offset in the trace stream.
36596 @item Packet opcode and payload
36600 @kindex maint btrace clear-packet-history
36601 @item maint btrace clear-packet-history
36602 Discards the cached packet history printed by the @samp{maint btrace
36603 packet-history} command. The history will be computed again when
36606 @kindex maint btrace clear
36607 @item maint btrace clear
36608 Discard the branch trace data. The data will be fetched anew and the
36609 branch trace will be recomputed when needed.
36611 This implicitly truncates the branch trace to a single branch trace
36612 buffer. When updating branch trace incrementally, the branch trace
36613 available to @value{GDBN} may be bigger than a single branch trace
36616 @kindex maint set btrace pt skip-pad
36617 @item maint set btrace pt skip-pad
36618 @kindex maint show btrace pt skip-pad
36619 @item maint show btrace pt skip-pad
36620 Control whether @value{GDBN} will skip PAD packets when computing the
36623 @kindex set displaced-stepping
36624 @kindex show displaced-stepping
36625 @cindex displaced stepping support
36626 @cindex out-of-line single-stepping
36627 @item set displaced-stepping
36628 @itemx show displaced-stepping
36629 Control whether or not @value{GDBN} will do @dfn{displaced stepping}
36630 if the target supports it. Displaced stepping is a way to single-step
36631 over breakpoints without removing them from the inferior, by executing
36632 an out-of-line copy of the instruction that was originally at the
36633 breakpoint location. It is also known as out-of-line single-stepping.
36636 @item set displaced-stepping on
36637 If the target architecture supports it, @value{GDBN} will use
36638 displaced stepping to step over breakpoints.
36640 @item set displaced-stepping off
36641 @value{GDBN} will not use displaced stepping to step over breakpoints,
36642 even if such is supported by the target architecture.
36644 @cindex non-stop mode, and @samp{set displaced-stepping}
36645 @item set displaced-stepping auto
36646 This is the default mode. @value{GDBN} will use displaced stepping
36647 only if non-stop mode is active (@pxref{Non-Stop Mode}) and the target
36648 architecture supports displaced stepping.
36651 @kindex maint check-psymtabs
36652 @item maint check-psymtabs
36653 Check the consistency of currently expanded psymtabs versus symtabs.
36654 Use this to check, for example, whether a symbol is in one but not the other.
36656 @kindex maint check-symtabs
36657 @item maint check-symtabs
36658 Check the consistency of currently expanded symtabs.
36660 @kindex maint expand-symtabs
36661 @item maint expand-symtabs [@var{regexp}]
36662 Expand symbol tables.
36663 If @var{regexp} is specified, only expand symbol tables for file
36664 names matching @var{regexp}.
36666 @kindex maint set catch-demangler-crashes
36667 @kindex maint show catch-demangler-crashes
36668 @cindex demangler crashes
36669 @item maint set catch-demangler-crashes [on|off]
36670 @itemx maint show catch-demangler-crashes
36671 Control whether @value{GDBN} should attempt to catch crashes in the
36672 symbol name demangler. The default is to attempt to catch crashes.
36673 If enabled, the first time a crash is caught, a core file is created,
36674 the offending symbol is displayed and the user is presented with the
36675 option to terminate the current session.
36677 @kindex maint cplus first_component
36678 @item maint cplus first_component @var{name}
36679 Print the first C@t{++} class/namespace component of @var{name}.
36681 @kindex maint cplus namespace
36682 @item maint cplus namespace
36683 Print the list of possible C@t{++} namespaces.
36685 @kindex maint deprecate
36686 @kindex maint undeprecate
36687 @cindex deprecated commands
36688 @item maint deprecate @var{command} @r{[}@var{replacement}@r{]}
36689 @itemx maint undeprecate @var{command}
36690 Deprecate or undeprecate the named @var{command}. Deprecated commands
36691 cause @value{GDBN} to issue a warning when you use them. The optional
36692 argument @var{replacement} says which newer command should be used in
36693 favor of the deprecated one; if it is given, @value{GDBN} will mention
36694 the replacement as part of the warning.
36696 @kindex maint dump-me
36697 @item maint dump-me
36698 @cindex @code{SIGQUIT} signal, dump core of @value{GDBN}
36699 Cause a fatal signal in the debugger and force it to dump its core.
36700 This is supported only on systems which support aborting a program
36701 with the @code{SIGQUIT} signal.
36703 @kindex maint internal-error
36704 @kindex maint internal-warning
36705 @kindex maint demangler-warning
36706 @cindex demangler crashes
36707 @item maint internal-error @r{[}@var{message-text}@r{]}
36708 @itemx maint internal-warning @r{[}@var{message-text}@r{]}
36709 @itemx maint demangler-warning @r{[}@var{message-text}@r{]}
36711 Cause @value{GDBN} to call the internal function @code{internal_error},
36712 @code{internal_warning} or @code{demangler_warning} and hence behave
36713 as though an internal problem has been detected. In addition to
36714 reporting the internal problem, these functions give the user the
36715 opportunity to either quit @value{GDBN} or (for @code{internal_error}
36716 and @code{internal_warning}) create a core file of the current
36717 @value{GDBN} session.
36719 These commands take an optional parameter @var{message-text} that is
36720 used as the text of the error or warning message.
36722 Here's an example of using @code{internal-error}:
36725 (@value{GDBP}) @kbd{maint internal-error testing, 1, 2}
36726 @dots{}/maint.c:121: internal-error: testing, 1, 2
36727 A problem internal to GDB has been detected. Further
36728 debugging may prove unreliable.
36729 Quit this debugging session? (y or n) @kbd{n}
36730 Create a core file? (y or n) @kbd{n}
36734 @cindex @value{GDBN} internal error
36735 @cindex internal errors, control of @value{GDBN} behavior
36736 @cindex demangler crashes
36738 @kindex maint set internal-error
36739 @kindex maint show internal-error
36740 @kindex maint set internal-warning
36741 @kindex maint show internal-warning
36742 @kindex maint set demangler-warning
36743 @kindex maint show demangler-warning
36744 @item maint set internal-error @var{action} [ask|yes|no]
36745 @itemx maint show internal-error @var{action}
36746 @itemx maint set internal-warning @var{action} [ask|yes|no]
36747 @itemx maint show internal-warning @var{action}
36748 @itemx maint set demangler-warning @var{action} [ask|yes|no]
36749 @itemx maint show demangler-warning @var{action}
36750 When @value{GDBN} reports an internal problem (error or warning) it
36751 gives the user the opportunity to both quit @value{GDBN} and create a
36752 core file of the current @value{GDBN} session. These commands let you
36753 override the default behaviour for each particular @var{action},
36754 described in the table below.
36758 You can specify that @value{GDBN} should always (yes) or never (no)
36759 quit. The default is to ask the user what to do.
36762 You can specify that @value{GDBN} should always (yes) or never (no)
36763 create a core file. The default is to ask the user what to do. Note
36764 that there is no @code{corefile} option for @code{demangler-warning}:
36765 demangler warnings always create a core file and this cannot be
36769 @kindex maint packet
36770 @item maint packet @var{text}
36771 If @value{GDBN} is talking to an inferior via the serial protocol,
36772 then this command sends the string @var{text} to the inferior, and
36773 displays the response packet. @value{GDBN} supplies the initial
36774 @samp{$} character, the terminating @samp{#} character, and the
36777 @kindex maint print architecture
36778 @item maint print architecture @r{[}@var{file}@r{]}
36779 Print the entire architecture configuration. The optional argument
36780 @var{file} names the file where the output goes.
36782 @kindex maint print c-tdesc @r{[}@var{file}@r{]}
36783 @item maint print c-tdesc
36784 Print the target description (@pxref{Target Descriptions}) as
36785 a C source file. By default, the target description is for the current
36786 target, but if the optional argument @var{file} is provided, that file
36787 is used to produce the description. The @var{file} should be an XML
36788 document, of the form described in @ref{Target Description Format}.
36789 The created source file is built into @value{GDBN} when @value{GDBN} is
36790 built again. This command is used by developers after they add or
36791 modify XML target descriptions.
36793 @kindex maint check xml-descriptions
36794 @item maint check xml-descriptions @var{dir}
36795 Check that the target descriptions dynamically created by @value{GDBN}
36796 equal the descriptions created from XML files found in @var{dir}.
36798 @anchor{maint check libthread-db}
36799 @kindex maint check libthread-db
36800 @item maint check libthread-db
36801 Run integrity checks on the current inferior's thread debugging
36802 library. This exercises all @code{libthread_db} functionality used by
36803 @value{GDBN} on GNU/Linux systems, and by extension also exercises the
36804 @code{proc_service} functions provided by @value{GDBN} that
36805 @code{libthread_db} uses. Note that parts of the test may be skipped
36806 on some platforms when debugging core files.
36808 @kindex maint print dummy-frames
36809 @item maint print dummy-frames
36810 Prints the contents of @value{GDBN}'s internal dummy-frame stack.
36813 (@value{GDBP}) @kbd{b add}
36815 (@value{GDBP}) @kbd{print add(2,3)}
36816 Breakpoint 2, add (a=2, b=3) at @dots{}
36818 The program being debugged stopped while in a function called from GDB.
36820 (@value{GDBP}) @kbd{maint print dummy-frames}
36821 0xa8206d8: id=@{stack=0xbfffe734,code=0xbfffe73f,!special@}, ptid=process 9353
36825 Takes an optional file parameter.
36827 @kindex maint print registers
36828 @kindex maint print raw-registers
36829 @kindex maint print cooked-registers
36830 @kindex maint print register-groups
36831 @kindex maint print remote-registers
36832 @item maint print registers @r{[}@var{file}@r{]}
36833 @itemx maint print raw-registers @r{[}@var{file}@r{]}
36834 @itemx maint print cooked-registers @r{[}@var{file}@r{]}
36835 @itemx maint print register-groups @r{[}@var{file}@r{]}
36836 @itemx maint print remote-registers @r{[}@var{file}@r{]}
36837 Print @value{GDBN}'s internal register data structures.
36839 The command @code{maint print raw-registers} includes the contents of
36840 the raw register cache; the command @code{maint print
36841 cooked-registers} includes the (cooked) value of all registers,
36842 including registers which aren't available on the target nor visible
36843 to user; the command @code{maint print register-groups} includes the
36844 groups that each register is a member of; and the command @code{maint
36845 print remote-registers} includes the remote target's register numbers
36846 and offsets in the `G' packets.
36848 These commands take an optional parameter, a file name to which to
36849 write the information.
36851 @kindex maint print reggroups
36852 @item maint print reggroups @r{[}@var{file}@r{]}
36853 Print @value{GDBN}'s internal register group data structures. The
36854 optional argument @var{file} tells to what file to write the
36857 The register groups info looks like this:
36860 (@value{GDBP}) @kbd{maint print reggroups}
36873 This command forces @value{GDBN} to flush its internal register cache.
36875 @kindex maint print objfiles
36876 @cindex info for known object files
36877 @item maint print objfiles @r{[}@var{regexp}@r{]}
36878 Print a dump of all known object files.
36879 If @var{regexp} is specified, only print object files whose names
36880 match @var{regexp}. For each object file, this command prints its name,
36881 address in memory, and all of its psymtabs and symtabs.
36883 @kindex maint print user-registers
36884 @cindex user registers
36885 @item maint print user-registers
36886 List all currently available @dfn{user registers}. User registers
36887 typically provide alternate names for actual hardware registers. They
36888 include the four ``standard'' registers @code{$fp}, @code{$pc},
36889 @code{$sp}, and @code{$ps}. @xref{standard registers}. User
36890 registers can be used in expressions in the same way as the canonical
36891 register names, but only the latter are listed by the @code{info
36892 registers} and @code{maint print registers} commands.
36894 @kindex maint print section-scripts
36895 @cindex info for known .debug_gdb_scripts-loaded scripts
36896 @item maint print section-scripts [@var{regexp}]
36897 Print a dump of scripts specified in the @code{.debug_gdb_section} section.
36898 If @var{regexp} is specified, only print scripts loaded by object files
36899 matching @var{regexp}.
36900 For each script, this command prints its name as specified in the objfile,
36901 and the full path if known.
36902 @xref{dotdebug_gdb_scripts section}.
36904 @kindex maint print statistics
36905 @cindex bcache statistics
36906 @item maint print statistics
36907 This command prints, for each object file in the program, various data
36908 about that object file followed by the byte cache (@dfn{bcache})
36909 statistics for the object file. The objfile data includes the number
36910 of minimal, partial, full, and stabs symbols, the number of types
36911 defined by the objfile, the number of as yet unexpanded psym tables,
36912 the number of line tables and string tables, and the amount of memory
36913 used by the various tables. The bcache statistics include the counts,
36914 sizes, and counts of duplicates of all and unique objects, max,
36915 average, and median entry size, total memory used and its overhead and
36916 savings, and various measures of the hash table size and chain
36919 @kindex maint print target-stack
36920 @cindex target stack description
36921 @item maint print target-stack
36922 A @dfn{target} is an interface between the debugger and a particular
36923 kind of file or process. Targets can be stacked in @dfn{strata},
36924 so that more than one target can potentially respond to a request.
36925 In particular, memory accesses will walk down the stack of targets
36926 until they find a target that is interested in handling that particular
36929 This command prints a short description of each layer that was pushed on
36930 the @dfn{target stack}, starting from the top layer down to the bottom one.
36932 @kindex maint print type
36933 @cindex type chain of a data type
36934 @item maint print type @var{expr}
36935 Print the type chain for a type specified by @var{expr}. The argument
36936 can be either a type name or a symbol. If it is a symbol, the type of
36937 that symbol is described. The type chain produced by this command is
36938 a recursive definition of the data type as stored in @value{GDBN}'s
36939 data structures, including its flags and contained types.
36941 @kindex maint selftest
36943 @item maint selftest @r{[}@var{filter}@r{]}
36944 Run any self tests that were compiled in to @value{GDBN}. This will
36945 print a message showing how many tests were run, and how many failed.
36946 If a @var{filter} is passed, only the tests with @var{filter} in their
36949 @kindex maint info selftests
36951 @item maint info selftests
36952 List the selftests compiled in to @value{GDBN}.
36954 @kindex maint set dwarf always-disassemble
36955 @kindex maint show dwarf always-disassemble
36956 @item maint set dwarf always-disassemble
36957 @item maint show dwarf always-disassemble
36958 Control the behavior of @code{info address} when using DWARF debugging
36961 The default is @code{off}, which means that @value{GDBN} should try to
36962 describe a variable's location in an easily readable format. When
36963 @code{on}, @value{GDBN} will instead display the DWARF location
36964 expression in an assembly-like format. Note that some locations are
36965 too complex for @value{GDBN} to describe simply; in this case you will
36966 always see the disassembly form.
36968 Here is an example of the resulting disassembly:
36971 (gdb) info addr argc
36972 Symbol "argc" is a complex DWARF expression:
36976 For more information on these expressions, see
36977 @uref{http://www.dwarfstd.org/, the DWARF standard}.
36979 @kindex maint set dwarf max-cache-age
36980 @kindex maint show dwarf max-cache-age
36981 @item maint set dwarf max-cache-age
36982 @itemx maint show dwarf max-cache-age
36983 Control the DWARF compilation unit cache.
36985 @cindex DWARF compilation units cache
36986 In object files with inter-compilation-unit references, such as those
36987 produced by the GCC option @samp{-feliminate-dwarf2-dups}, the DWARF
36988 reader needs to frequently refer to previously read compilation units.
36989 This setting controls how long a compilation unit will remain in the
36990 cache if it is not referenced. A higher limit means that cached
36991 compilation units will be stored in memory longer, and more total
36992 memory will be used. Setting it to zero disables caching, which will
36993 slow down @value{GDBN} startup, but reduce memory consumption.
36995 @kindex maint set dwarf unwinders
36996 @kindex maint show dwarf unwinders
36997 @item maint set dwarf unwinders
36998 @itemx maint show dwarf unwinders
36999 Control use of the DWARF frame unwinders.
37001 @cindex DWARF frame unwinders
37002 Many targets that support DWARF debugging use @value{GDBN}'s DWARF
37003 frame unwinders to build the backtrace. Many of these targets will
37004 also have a second mechanism for building the backtrace for use in
37005 cases where DWARF information is not available, this second mechanism
37006 is often an analysis of a function's prologue.
37008 In order to extend testing coverage of the second level stack
37009 unwinding mechanisms it is helpful to be able to disable the DWARF
37010 stack unwinders, this can be done with this switch.
37012 In normal use of @value{GDBN} disabling the DWARF unwinders is not
37013 advisable, there are cases that are better handled through DWARF than
37014 prologue analysis, and the debug experience is likely to be better
37015 with the DWARF frame unwinders enabled.
37017 If DWARF frame unwinders are not supported for a particular target
37018 architecture, then enabling this flag does not cause them to be used.
37019 @kindex maint set profile
37020 @kindex maint show profile
37021 @cindex profiling GDB
37022 @item maint set profile
37023 @itemx maint show profile
37024 Control profiling of @value{GDBN}.
37026 Profiling will be disabled until you use the @samp{maint set profile}
37027 command to enable it. When you enable profiling, the system will begin
37028 collecting timing and execution count data; when you disable profiling or
37029 exit @value{GDBN}, the results will be written to a log file. Remember that
37030 if you use profiling, @value{GDBN} will overwrite the profiling log file
37031 (often called @file{gmon.out}). If you have a record of important profiling
37032 data in a @file{gmon.out} file, be sure to move it to a safe location.
37034 Configuring with @samp{--enable-profiling} arranges for @value{GDBN} to be
37035 compiled with the @samp{-pg} compiler option.
37037 @kindex maint set show-debug-regs
37038 @kindex maint show show-debug-regs
37039 @cindex hardware debug registers
37040 @item maint set show-debug-regs
37041 @itemx maint show show-debug-regs
37042 Control whether to show variables that mirror the hardware debug
37043 registers. Use @code{on} to enable, @code{off} to disable. If
37044 enabled, the debug registers values are shown when @value{GDBN} inserts or
37045 removes a hardware breakpoint or watchpoint, and when the inferior
37046 triggers a hardware-assisted breakpoint or watchpoint.
37048 @kindex maint set show-all-tib
37049 @kindex maint show show-all-tib
37050 @item maint set show-all-tib
37051 @itemx maint show show-all-tib
37052 Control whether to show all non zero areas within a 1k block starting
37053 at thread local base, when using the @samp{info w32 thread-information-block}
37056 @kindex maint set target-async
37057 @kindex maint show target-async
37058 @item maint set target-async
37059 @itemx maint show target-async
37060 This controls whether @value{GDBN} targets operate in synchronous or
37061 asynchronous mode (@pxref{Background Execution}). Normally the
37062 default is asynchronous, if it is available; but this can be changed
37063 to more easily debug problems occurring only in synchronous mode.
37065 @kindex maint set target-non-stop @var{mode} [on|off|auto]
37066 @kindex maint show target-non-stop
37067 @item maint set target-non-stop
37068 @itemx maint show target-non-stop
37070 This controls whether @value{GDBN} targets always operate in non-stop
37071 mode even if @code{set non-stop} is @code{off} (@pxref{Non-Stop
37072 Mode}). The default is @code{auto}, meaning non-stop mode is enabled
37073 if supported by the target.
37076 @item maint set target-non-stop auto
37077 This is the default mode. @value{GDBN} controls the target in
37078 non-stop mode if the target supports it.
37080 @item maint set target-non-stop on
37081 @value{GDBN} controls the target in non-stop mode even if the target
37082 does not indicate support.
37084 @item maint set target-non-stop off
37085 @value{GDBN} does not control the target in non-stop mode even if the
37086 target supports it.
37089 @kindex maint set per-command
37090 @kindex maint show per-command
37091 @item maint set per-command
37092 @itemx maint show per-command
37093 @cindex resources used by commands
37095 @value{GDBN} can display the resources used by each command.
37096 This is useful in debugging performance problems.
37099 @item maint set per-command space [on|off]
37100 @itemx maint show per-command space
37101 Enable or disable the printing of the memory used by GDB for each command.
37102 If enabled, @value{GDBN} will display how much memory each command
37103 took, following the command's own output.
37104 This can also be requested by invoking @value{GDBN} with the
37105 @option{--statistics} command-line switch (@pxref{Mode Options}).
37107 @item maint set per-command time [on|off]
37108 @itemx maint show per-command time
37109 Enable or disable the printing of the execution time of @value{GDBN}
37111 If enabled, @value{GDBN} will display how much time it
37112 took to execute each command, following the command's own output.
37113 Both CPU time and wallclock time are printed.
37114 Printing both is useful when trying to determine whether the cost is
37115 CPU or, e.g., disk/network latency.
37116 Note that the CPU time printed is for @value{GDBN} only, it does not include
37117 the execution time of the inferior because there's no mechanism currently
37118 to compute how much time was spent by @value{GDBN} and how much time was
37119 spent by the program been debugged.
37120 This can also be requested by invoking @value{GDBN} with the
37121 @option{--statistics} command-line switch (@pxref{Mode Options}).
37123 @item maint set per-command symtab [on|off]
37124 @itemx maint show per-command symtab
37125 Enable or disable the printing of basic symbol table statistics
37127 If enabled, @value{GDBN} will display the following information:
37131 number of symbol tables
37133 number of primary symbol tables
37135 number of blocks in the blockvector
37139 @kindex maint set check-libthread-db
37140 @kindex maint show check-libthread-db
37141 @item maint set check-libthread-db [on|off]
37142 @itemx maint show check-libthread-db
37143 Control whether @value{GDBN} should run integrity checks on inferior
37144 specific thread debugging libraries as they are loaded. The default
37145 is not to perform such checks. If any check fails @value{GDBN} will
37146 unload the library and continue searching for a suitable candidate as
37147 described in @ref{set libthread-db-search-path}. For more information
37148 about the tests, see @ref{maint check libthread-db}.
37150 @kindex maint space
37151 @cindex memory used by commands
37152 @item maint space @var{value}
37153 An alias for @code{maint set per-command space}.
37154 A non-zero value enables it, zero disables it.
37157 @cindex time of command execution
37158 @item maint time @var{value}
37159 An alias for @code{maint set per-command time}.
37160 A non-zero value enables it, zero disables it.
37162 @kindex maint translate-address
37163 @item maint translate-address @r{[}@var{section}@r{]} @var{addr}
37164 Find the symbol stored at the location specified by the address
37165 @var{addr} and an optional section name @var{section}. If found,
37166 @value{GDBN} prints the name of the closest symbol and an offset from
37167 the symbol's location to the specified address. This is similar to
37168 the @code{info address} command (@pxref{Symbols}), except that this
37169 command also allows to find symbols in other sections.
37171 If section was not specified, the section in which the symbol was found
37172 is also printed. For dynamically linked executables, the name of
37173 executable or shared library containing the symbol is printed as well.
37177 The following command is useful for non-interactive invocations of
37178 @value{GDBN}, such as in the test suite.
37181 @item set watchdog @var{nsec}
37182 @kindex set watchdog
37183 @cindex watchdog timer
37184 @cindex timeout for commands
37185 Set the maximum number of seconds @value{GDBN} will wait for the
37186 target operation to finish. If this time expires, @value{GDBN}
37187 reports and error and the command is aborted.
37189 @item show watchdog
37190 Show the current setting of the target wait timeout.
37193 @node Remote Protocol
37194 @appendix @value{GDBN} Remote Serial Protocol
37199 * Stop Reply Packets::
37200 * General Query Packets::
37201 * Architecture-Specific Protocol Details::
37202 * Tracepoint Packets::
37203 * Host I/O Packets::
37205 * Notification Packets::
37206 * Remote Non-Stop::
37207 * Packet Acknowledgment::
37209 * File-I/O Remote Protocol Extension::
37210 * Library List Format::
37211 * Library List Format for SVR4 Targets::
37212 * Memory Map Format::
37213 * Thread List Format::
37214 * Traceframe Info Format::
37215 * Branch Trace Format::
37216 * Branch Trace Configuration Format::
37222 There may be occasions when you need to know something about the
37223 protocol---for example, if there is only one serial port to your target
37224 machine, you might want your program to do something special if it
37225 recognizes a packet meant for @value{GDBN}.
37227 In the examples below, @samp{->} and @samp{<-} are used to indicate
37228 transmitted and received data, respectively.
37230 @cindex protocol, @value{GDBN} remote serial
37231 @cindex serial protocol, @value{GDBN} remote
37232 @cindex remote serial protocol
37233 All @value{GDBN} commands and responses (other than acknowledgments
37234 and notifications, see @ref{Notification Packets}) are sent as a
37235 @var{packet}. A @var{packet} is introduced with the character
37236 @samp{$}, the actual @var{packet-data}, and the terminating character
37237 @samp{#} followed by a two-digit @var{checksum}:
37240 @code{$}@var{packet-data}@code{#}@var{checksum}
37244 @cindex checksum, for @value{GDBN} remote
37246 The two-digit @var{checksum} is computed as the modulo 256 sum of all
37247 characters between the leading @samp{$} and the trailing @samp{#} (an
37248 eight bit unsigned checksum).
37250 Implementors should note that prior to @value{GDBN} 5.0 the protocol
37251 specification also included an optional two-digit @var{sequence-id}:
37254 @code{$}@var{sequence-id}@code{:}@var{packet-data}@code{#}@var{checksum}
37257 @cindex sequence-id, for @value{GDBN} remote
37259 That @var{sequence-id} was appended to the acknowledgment. @value{GDBN}
37260 has never output @var{sequence-id}s. Stubs that handle packets added
37261 since @value{GDBN} 5.0 must not accept @var{sequence-id}.
37263 When either the host or the target machine receives a packet, the first
37264 response expected is an acknowledgment: either @samp{+} (to indicate
37265 the package was received correctly) or @samp{-} (to request
37269 -> @code{$}@var{packet-data}@code{#}@var{checksum}
37274 The @samp{+}/@samp{-} acknowledgments can be disabled
37275 once a connection is established.
37276 @xref{Packet Acknowledgment}, for details.
37278 The host (@value{GDBN}) sends @var{command}s, and the target (the
37279 debugging stub incorporated in your program) sends a @var{response}. In
37280 the case of step and continue @var{command}s, the response is only sent
37281 when the operation has completed, and the target has again stopped all
37282 threads in all attached processes. This is the default all-stop mode
37283 behavior, but the remote protocol also supports @value{GDBN}'s non-stop
37284 execution mode; see @ref{Remote Non-Stop}, for details.
37286 @var{packet-data} consists of a sequence of characters with the
37287 exception of @samp{#} and @samp{$} (see @samp{X} packet for additional
37290 @cindex remote protocol, field separator
37291 Fields within the packet should be separated using @samp{,} @samp{;} or
37292 @samp{:}. Except where otherwise noted all numbers are represented in
37293 @sc{hex} with leading zeros suppressed.
37295 Implementors should note that prior to @value{GDBN} 5.0, the character
37296 @samp{:} could not appear as the third character in a packet (as it
37297 would potentially conflict with the @var{sequence-id}).
37299 @cindex remote protocol, binary data
37300 @anchor{Binary Data}
37301 Binary data in most packets is encoded either as two hexadecimal
37302 digits per byte of binary data. This allowed the traditional remote
37303 protocol to work over connections which were only seven-bit clean.
37304 Some packets designed more recently assume an eight-bit clean
37305 connection, and use a more efficient encoding to send and receive
37308 The binary data representation uses @code{7d} (@sc{ascii} @samp{@}})
37309 as an escape character. Any escaped byte is transmitted as the escape
37310 character followed by the original character XORed with @code{0x20}.
37311 For example, the byte @code{0x7d} would be transmitted as the two
37312 bytes @code{0x7d 0x5d}. The bytes @code{0x23} (@sc{ascii} @samp{#}),
37313 @code{0x24} (@sc{ascii} @samp{$}), and @code{0x7d} (@sc{ascii}
37314 @samp{@}}) must always be escaped. Responses sent by the stub
37315 must also escape @code{0x2a} (@sc{ascii} @samp{*}), so that it
37316 is not interpreted as the start of a run-length encoded sequence
37319 Response @var{data} can be run-length encoded to save space.
37320 Run-length encoding replaces runs of identical characters with one
37321 instance of the repeated character, followed by a @samp{*} and a
37322 repeat count. The repeat count is itself sent encoded, to avoid
37323 binary characters in @var{data}: a value of @var{n} is sent as
37324 @code{@var{n}+29}. For a repeat count greater or equal to 3, this
37325 produces a printable @sc{ascii} character, e.g.@: a space (@sc{ascii}
37326 code 32) for a repeat count of 3. (This is because run-length
37327 encoding starts to win for counts 3 or more.) Thus, for example,
37328 @samp{0* } is a run-length encoding of ``0000'': the space character
37329 after @samp{*} means repeat the leading @code{0} @w{@code{32 - 29 =
37332 The printable characters @samp{#} and @samp{$} or with a numeric value
37333 greater than 126 must not be used. Runs of six repeats (@samp{#}) or
37334 seven repeats (@samp{$}) can be expanded using a repeat count of only
37335 five (@samp{"}). For example, @samp{00000000} can be encoded as
37338 The error response returned for some packets includes a two character
37339 error number. That number is not well defined.
37341 @cindex empty response, for unsupported packets
37342 For any @var{command} not supported by the stub, an empty response
37343 (@samp{$#00}) should be returned. That way it is possible to extend the
37344 protocol. A newer @value{GDBN} can tell if a packet is supported based
37347 At a minimum, a stub is required to support the @samp{g} and @samp{G}
37348 commands for register access, and the @samp{m} and @samp{M} commands
37349 for memory access. Stubs that only control single-threaded targets
37350 can implement run control with the @samp{c} (continue), and @samp{s}
37351 (step) commands. Stubs that support multi-threading targets should
37352 support the @samp{vCont} command. All other commands are optional.
37357 The following table provides a complete list of all currently defined
37358 @var{command}s and their corresponding response @var{data}.
37359 @xref{File-I/O Remote Protocol Extension}, for details about the File
37360 I/O extension of the remote protocol.
37362 Each packet's description has a template showing the packet's overall
37363 syntax, followed by an explanation of the packet's meaning. We
37364 include spaces in some of the templates for clarity; these are not
37365 part of the packet's syntax. No @value{GDBN} packet uses spaces to
37366 separate its components. For example, a template like @samp{foo
37367 @var{bar} @var{baz}} describes a packet beginning with the three ASCII
37368 bytes @samp{foo}, followed by a @var{bar}, followed directly by a
37369 @var{baz}. @value{GDBN} does not transmit a space character between the
37370 @samp{foo} and the @var{bar}, or between the @var{bar} and the
37373 @cindex @var{thread-id}, in remote protocol
37374 @anchor{thread-id syntax}
37375 Several packets and replies include a @var{thread-id} field to identify
37376 a thread. Normally these are positive numbers with a target-specific
37377 interpretation, formatted as big-endian hex strings. A @var{thread-id}
37378 can also be a literal @samp{-1} to indicate all threads, or @samp{0} to
37381 In addition, the remote protocol supports a multiprocess feature in
37382 which the @var{thread-id} syntax is extended to optionally include both
37383 process and thread ID fields, as @samp{p@var{pid}.@var{tid}}.
37384 The @var{pid} (process) and @var{tid} (thread) components each have the
37385 format described above: a positive number with target-specific
37386 interpretation formatted as a big-endian hex string, literal @samp{-1}
37387 to indicate all processes or threads (respectively), or @samp{0} to
37388 indicate an arbitrary process or thread. Specifying just a process, as
37389 @samp{p@var{pid}}, is equivalent to @samp{p@var{pid}.-1}. It is an
37390 error to specify all processes but a specific thread, such as
37391 @samp{p-1.@var{tid}}. Note that the @samp{p} prefix is @emph{not} used
37392 for those packets and replies explicitly documented to include a process
37393 ID, rather than a @var{thread-id}.
37395 The multiprocess @var{thread-id} syntax extensions are only used if both
37396 @value{GDBN} and the stub report support for the @samp{multiprocess}
37397 feature using @samp{qSupported}. @xref{multiprocess extensions}, for
37400 Note that all packet forms beginning with an upper- or lower-case
37401 letter, other than those described here, are reserved for future use.
37403 Here are the packet descriptions.
37408 @cindex @samp{!} packet
37409 @anchor{extended mode}
37410 Enable extended mode. In extended mode, the remote server is made
37411 persistent. The @samp{R} packet is used to restart the program being
37417 The remote target both supports and has enabled extended mode.
37421 @cindex @samp{?} packet
37423 Indicate the reason the target halted. The reply is the same as for
37424 step and continue. This packet has a special interpretation when the
37425 target is in non-stop mode; see @ref{Remote Non-Stop}.
37428 @xref{Stop Reply Packets}, for the reply specifications.
37430 @item A @var{arglen},@var{argnum},@var{arg},@dots{}
37431 @cindex @samp{A} packet
37432 Initialized @code{argv[]} array passed into program. @var{arglen}
37433 specifies the number of bytes in the hex encoded byte stream
37434 @var{arg}. See @code{gdbserver} for more details.
37439 The arguments were set.
37445 @cindex @samp{b} packet
37446 (Don't use this packet; its behavior is not well-defined.)
37447 Change the serial line speed to @var{baud}.
37449 JTC: @emph{When does the transport layer state change? When it's
37450 received, or after the ACK is transmitted. In either case, there are
37451 problems if the command or the acknowledgment packet is dropped.}
37453 Stan: @emph{If people really wanted to add something like this, and get
37454 it working for the first time, they ought to modify ser-unix.c to send
37455 some kind of out-of-band message to a specially-setup stub and have the
37456 switch happen "in between" packets, so that from remote protocol's point
37457 of view, nothing actually happened.}
37459 @item B @var{addr},@var{mode}
37460 @cindex @samp{B} packet
37461 Set (@var{mode} is @samp{S}) or clear (@var{mode} is @samp{C}) a
37462 breakpoint at @var{addr}.
37464 Don't use this packet. Use the @samp{Z} and @samp{z} packets instead
37465 (@pxref{insert breakpoint or watchpoint packet}).
37467 @cindex @samp{bc} packet
37470 Backward continue. Execute the target system in reverse. No parameter.
37471 @xref{Reverse Execution}, for more information.
37474 @xref{Stop Reply Packets}, for the reply specifications.
37476 @cindex @samp{bs} packet
37479 Backward single step. Execute one instruction in reverse. No parameter.
37480 @xref{Reverse Execution}, for more information.
37483 @xref{Stop Reply Packets}, for the reply specifications.
37485 @item c @r{[}@var{addr}@r{]}
37486 @cindex @samp{c} packet
37487 Continue at @var{addr}, which is the address to resume. If @var{addr}
37488 is omitted, resume at current address.
37490 This packet is deprecated for multi-threading support. @xref{vCont
37494 @xref{Stop Reply Packets}, for the reply specifications.
37496 @item C @var{sig}@r{[};@var{addr}@r{]}
37497 @cindex @samp{C} packet
37498 Continue with signal @var{sig} (hex signal number). If
37499 @samp{;@var{addr}} is omitted, resume at same address.
37501 This packet is deprecated for multi-threading support. @xref{vCont
37505 @xref{Stop Reply Packets}, for the reply specifications.
37508 @cindex @samp{d} packet
37511 Don't use this packet; instead, define a general set packet
37512 (@pxref{General Query Packets}).
37516 @cindex @samp{D} packet
37517 The first form of the packet is used to detach @value{GDBN} from the
37518 remote system. It is sent to the remote target
37519 before @value{GDBN} disconnects via the @code{detach} command.
37521 The second form, including a process ID, is used when multiprocess
37522 protocol extensions are enabled (@pxref{multiprocess extensions}), to
37523 detach only a specific process. The @var{pid} is specified as a
37524 big-endian hex string.
37534 @item F @var{RC},@var{EE},@var{CF};@var{XX}
37535 @cindex @samp{F} packet
37536 A reply from @value{GDBN} to an @samp{F} packet sent by the target.
37537 This is part of the File-I/O protocol extension. @xref{File-I/O
37538 Remote Protocol Extension}, for the specification.
37541 @anchor{read registers packet}
37542 @cindex @samp{g} packet
37543 Read general registers.
37547 @item @var{XX@dots{}}
37548 Each byte of register data is described by two hex digits. The bytes
37549 with the register are transmitted in target byte order. The size of
37550 each register and their position within the @samp{g} packet are
37551 determined by the @value{GDBN} internal gdbarch functions
37552 @code{DEPRECATED_REGISTER_RAW_SIZE} and @code{gdbarch_register_name}.
37554 When reading registers from a trace frame (@pxref{Analyze Collected
37555 Data,,Using the Collected Data}), the stub may also return a string of
37556 literal @samp{x}'s in place of the register data digits, to indicate
37557 that the corresponding register has not been collected, thus its value
37558 is unavailable. For example, for an architecture with 4 registers of
37559 4 bytes each, the following reply indicates to @value{GDBN} that
37560 registers 0 and 2 have not been collected, while registers 1 and 3
37561 have been collected, and both have zero value:
37565 <- @code{xxxxxxxx00000000xxxxxxxx00000000}
37572 @item G @var{XX@dots{}}
37573 @cindex @samp{G} packet
37574 Write general registers. @xref{read registers packet}, for a
37575 description of the @var{XX@dots{}} data.
37585 @item H @var{op} @var{thread-id}
37586 @cindex @samp{H} packet
37587 Set thread for subsequent operations (@samp{m}, @samp{M}, @samp{g},
37588 @samp{G}, et.al.). Depending on the operation to be performed, @var{op}
37589 should be @samp{c} for step and continue operations (note that this
37590 is deprecated, supporting the @samp{vCont} command is a better
37591 option), and @samp{g} for other operations. The thread designator
37592 @var{thread-id} has the format and interpretation described in
37593 @ref{thread-id syntax}.
37604 @c 'H': How restrictive (or permissive) is the thread model. If a
37605 @c thread is selected and stopped, are other threads allowed
37606 @c to continue to execute? As I mentioned above, I think the
37607 @c semantics of each command when a thread is selected must be
37608 @c described. For example:
37610 @c 'g': If the stub supports threads and a specific thread is
37611 @c selected, returns the register block from that thread;
37612 @c otherwise returns current registers.
37614 @c 'G' If the stub supports threads and a specific thread is
37615 @c selected, sets the registers of the register block of
37616 @c that thread; otherwise sets current registers.
37618 @item i @r{[}@var{addr}@r{[},@var{nnn}@r{]]}
37619 @anchor{cycle step packet}
37620 @cindex @samp{i} packet
37621 Step the remote target by a single clock cycle. If @samp{,@var{nnn}} is
37622 present, cycle step @var{nnn} cycles. If @var{addr} is present, cycle
37623 step starting at that address.
37626 @cindex @samp{I} packet
37627 Signal, then cycle step. @xref{step with signal packet}. @xref{cycle
37631 @cindex @samp{k} packet
37634 The exact effect of this packet is not specified.
37636 For a bare-metal target, it may power cycle or reset the target
37637 system. For that reason, the @samp{k} packet has no reply.
37639 For a single-process target, it may kill that process if possible.
37641 A multiple-process target may choose to kill just one process, or all
37642 that are under @value{GDBN}'s control. For more precise control, use
37643 the vKill packet (@pxref{vKill packet}).
37645 If the target system immediately closes the connection in response to
37646 @samp{k}, @value{GDBN} does not consider the lack of packet
37647 acknowledgment to be an error, and assumes the kill was successful.
37649 If connected using @kbd{target extended-remote}, and the target does
37650 not close the connection in response to a kill request, @value{GDBN}
37651 probes the target state as if a new connection was opened
37652 (@pxref{? packet}).
37654 @item m @var{addr},@var{length}
37655 @cindex @samp{m} packet
37656 Read @var{length} addressable memory units starting at address @var{addr}
37657 (@pxref{addressable memory unit}). Note that @var{addr} may not be aligned to
37658 any particular boundary.
37660 The stub need not use any particular size or alignment when gathering
37661 data from memory for the response; even if @var{addr} is word-aligned
37662 and @var{length} is a multiple of the word size, the stub is free to
37663 use byte accesses, or not. For this reason, this packet may not be
37664 suitable for accessing memory-mapped I/O devices.
37665 @cindex alignment of remote memory accesses
37666 @cindex size of remote memory accesses
37667 @cindex memory, alignment and size of remote accesses
37671 @item @var{XX@dots{}}
37672 Memory contents; each byte is transmitted as a two-digit hexadecimal number.
37673 The reply may contain fewer addressable memory units than requested if the
37674 server was able to read only part of the region of memory.
37679 @item M @var{addr},@var{length}:@var{XX@dots{}}
37680 @cindex @samp{M} packet
37681 Write @var{length} addressable memory units starting at address @var{addr}
37682 (@pxref{addressable memory unit}). The data is given by @var{XX@dots{}}; each
37683 byte is transmitted as a two-digit hexadecimal number.
37690 for an error (this includes the case where only part of the data was
37695 @cindex @samp{p} packet
37696 Read the value of register @var{n}; @var{n} is in hex.
37697 @xref{read registers packet}, for a description of how the returned
37698 register value is encoded.
37702 @item @var{XX@dots{}}
37703 the register's value
37707 Indicating an unrecognized @var{query}.
37710 @item P @var{n@dots{}}=@var{r@dots{}}
37711 @anchor{write register packet}
37712 @cindex @samp{P} packet
37713 Write register @var{n@dots{}} with value @var{r@dots{}}. The register
37714 number @var{n} is in hexadecimal, and @var{r@dots{}} contains two hex
37715 digits for each byte in the register (target byte order).
37725 @item q @var{name} @var{params}@dots{}
37726 @itemx Q @var{name} @var{params}@dots{}
37727 @cindex @samp{q} packet
37728 @cindex @samp{Q} packet
37729 General query (@samp{q}) and set (@samp{Q}). These packets are
37730 described fully in @ref{General Query Packets}.
37733 @cindex @samp{r} packet
37734 Reset the entire system.
37736 Don't use this packet; use the @samp{R} packet instead.
37739 @cindex @samp{R} packet
37740 Restart the program being debugged. The @var{XX}, while needed, is ignored.
37741 This packet is only available in extended mode (@pxref{extended mode}).
37743 The @samp{R} packet has no reply.
37745 @item s @r{[}@var{addr}@r{]}
37746 @cindex @samp{s} packet
37747 Single step, resuming at @var{addr}. If
37748 @var{addr} is omitted, resume at same address.
37750 This packet is deprecated for multi-threading support. @xref{vCont
37754 @xref{Stop Reply Packets}, for the reply specifications.
37756 @item S @var{sig}@r{[};@var{addr}@r{]}
37757 @anchor{step with signal packet}
37758 @cindex @samp{S} packet
37759 Step with signal. This is analogous to the @samp{C} packet, but
37760 requests a single-step, rather than a normal resumption of execution.
37762 This packet is deprecated for multi-threading support. @xref{vCont
37766 @xref{Stop Reply Packets}, for the reply specifications.
37768 @item t @var{addr}:@var{PP},@var{MM}
37769 @cindex @samp{t} packet
37770 Search backwards starting at address @var{addr} for a match with pattern
37771 @var{PP} and mask @var{MM}, both of which are are 4 byte long.
37772 There must be at least 3 digits in @var{addr}.
37774 @item T @var{thread-id}
37775 @cindex @samp{T} packet
37776 Find out if the thread @var{thread-id} is alive. @xref{thread-id syntax}.
37781 thread is still alive
37787 Packets starting with @samp{v} are identified by a multi-letter name,
37788 up to the first @samp{;} or @samp{?} (or the end of the packet).
37790 @item vAttach;@var{pid}
37791 @cindex @samp{vAttach} packet
37792 Attach to a new process with the specified process ID @var{pid}.
37793 The process ID is a
37794 hexadecimal integer identifying the process. In all-stop mode, all
37795 threads in the attached process are stopped; in non-stop mode, it may be
37796 attached without being stopped if that is supported by the target.
37798 @c In non-stop mode, on a successful vAttach, the stub should set the
37799 @c current thread to a thread of the newly-attached process. After
37800 @c attaching, GDB queries for the attached process's thread ID with qC.
37801 @c Also note that, from a user perspective, whether or not the
37802 @c target is stopped on attach in non-stop mode depends on whether you
37803 @c use the foreground or background version of the attach command, not
37804 @c on what vAttach does; GDB does the right thing with respect to either
37805 @c stopping or restarting threads.
37807 This packet is only available in extended mode (@pxref{extended mode}).
37813 @item @r{Any stop packet}
37814 for success in all-stop mode (@pxref{Stop Reply Packets})
37816 for success in non-stop mode (@pxref{Remote Non-Stop})
37819 @item vCont@r{[};@var{action}@r{[}:@var{thread-id}@r{]]}@dots{}
37820 @cindex @samp{vCont} packet
37821 @anchor{vCont packet}
37822 Resume the inferior, specifying different actions for each thread.
37824 For each inferior thread, the leftmost action with a matching
37825 @var{thread-id} is applied. Threads that don't match any action
37826 remain in their current state. Thread IDs are specified using the
37827 syntax described in @ref{thread-id syntax}. If multiprocess
37828 extensions (@pxref{multiprocess extensions}) are supported, actions
37829 can be specified to match all threads in a process by using the
37830 @samp{p@var{pid}.-1} form of the @var{thread-id}. An action with no
37831 @var{thread-id} matches all threads. Specifying no actions is an
37834 Currently supported actions are:
37840 Continue with signal @var{sig}. The signal @var{sig} should be two hex digits.
37844 Step with signal @var{sig}. The signal @var{sig} should be two hex digits.
37847 @item r @var{start},@var{end}
37848 Step once, and then keep stepping as long as the thread stops at
37849 addresses between @var{start} (inclusive) and @var{end} (exclusive).
37850 The remote stub reports a stop reply when either the thread goes out
37851 of the range or is stopped due to an unrelated reason, such as hitting
37852 a breakpoint. @xref{range stepping}.
37854 If the range is empty (@var{start} == @var{end}), then the action
37855 becomes equivalent to the @samp{s} action. In other words,
37856 single-step once, and report the stop (even if the stepped instruction
37857 jumps to @var{start}).
37859 (A stop reply may be sent at any point even if the PC is still within
37860 the stepping range; for example, it is valid to implement this packet
37861 in a degenerate way as a single instruction step operation.)
37865 The optional argument @var{addr} normally associated with the
37866 @samp{c}, @samp{C}, @samp{s}, and @samp{S} packets is
37867 not supported in @samp{vCont}.
37869 The @samp{t} action is only relevant in non-stop mode
37870 (@pxref{Remote Non-Stop}) and may be ignored by the stub otherwise.
37871 A stop reply should be generated for any affected thread not already stopped.
37872 When a thread is stopped by means of a @samp{t} action,
37873 the corresponding stop reply should indicate that the thread has stopped with
37874 signal @samp{0}, regardless of whether the target uses some other signal
37875 as an implementation detail.
37877 The server must ignore @samp{c}, @samp{C}, @samp{s}, @samp{S}, and
37878 @samp{r} actions for threads that are already running. Conversely,
37879 the server must ignore @samp{t} actions for threads that are already
37882 @emph{Note:} In non-stop mode, a thread is considered running until
37883 @value{GDBN} acknowleges an asynchronous stop notification for it with
37884 the @samp{vStopped} packet (@pxref{Remote Non-Stop}).
37886 The stub must support @samp{vCont} if it reports support for
37887 multiprocess extensions (@pxref{multiprocess extensions}).
37890 @xref{Stop Reply Packets}, for the reply specifications.
37893 @cindex @samp{vCont?} packet
37894 Request a list of actions supported by the @samp{vCont} packet.
37898 @item vCont@r{[};@var{action}@dots{}@r{]}
37899 The @samp{vCont} packet is supported. Each @var{action} is a supported
37900 command in the @samp{vCont} packet.
37902 The @samp{vCont} packet is not supported.
37905 @anchor{vCtrlC packet}
37907 @cindex @samp{vCtrlC} packet
37908 Interrupt remote target as if a control-C was pressed on the remote
37909 terminal. This is the equivalent to reacting to the @code{^C}
37910 (@samp{\003}, the control-C character) character in all-stop mode
37911 while the target is running, except this works in non-stop mode.
37912 @xref{interrupting remote targets}, for more info on the all-stop
37923 @item vFile:@var{operation}:@var{parameter}@dots{}
37924 @cindex @samp{vFile} packet
37925 Perform a file operation on the target system. For details,
37926 see @ref{Host I/O Packets}.
37928 @item vFlashErase:@var{addr},@var{length}
37929 @cindex @samp{vFlashErase} packet
37930 Direct the stub to erase @var{length} bytes of flash starting at
37931 @var{addr}. The region may enclose any number of flash blocks, but
37932 its start and end must fall on block boundaries, as indicated by the
37933 flash block size appearing in the memory map (@pxref{Memory Map
37934 Format}). @value{GDBN} groups flash memory programming operations
37935 together, and sends a @samp{vFlashDone} request after each group; the
37936 stub is allowed to delay erase operation until the @samp{vFlashDone}
37937 packet is received.
37947 @item vFlashWrite:@var{addr}:@var{XX@dots{}}
37948 @cindex @samp{vFlashWrite} packet
37949 Direct the stub to write data to flash address @var{addr}. The data
37950 is passed in binary form using the same encoding as for the @samp{X}
37951 packet (@pxref{Binary Data}). The memory ranges specified by
37952 @samp{vFlashWrite} packets preceding a @samp{vFlashDone} packet must
37953 not overlap, and must appear in order of increasing addresses
37954 (although @samp{vFlashErase} packets for higher addresses may already
37955 have been received; the ordering is guaranteed only between
37956 @samp{vFlashWrite} packets). If a packet writes to an address that was
37957 neither erased by a preceding @samp{vFlashErase} packet nor by some other
37958 target-specific method, the results are unpredictable.
37966 for vFlashWrite addressing non-flash memory
37972 @cindex @samp{vFlashDone} packet
37973 Indicate to the stub that flash programming operation is finished.
37974 The stub is permitted to delay or batch the effects of a group of
37975 @samp{vFlashErase} and @samp{vFlashWrite} packets until a
37976 @samp{vFlashDone} packet is received. The contents of the affected
37977 regions of flash memory are unpredictable until the @samp{vFlashDone}
37978 request is completed.
37980 @item vKill;@var{pid}
37981 @cindex @samp{vKill} packet
37982 @anchor{vKill packet}
37983 Kill the process with the specified process ID @var{pid}, which is a
37984 hexadecimal integer identifying the process. This packet is used in
37985 preference to @samp{k} when multiprocess protocol extensions are
37986 supported; see @ref{multiprocess extensions}.
37996 @item vMustReplyEmpty
37997 @cindex @samp{vMustReplyEmpty} packet
37998 The correct reply to an unknown @samp{v} packet is to return the empty
37999 string, however, some older versions of @command{gdbserver} would
38000 incorrectly return @samp{OK} for unknown @samp{v} packets.
38002 The @samp{vMustReplyEmpty} is used as a feature test to check how
38003 @command{gdbserver} handles unknown packets, it is important that this
38004 packet be handled in the same way as other unknown @samp{v} packets.
38005 If this packet is handled differently to other unknown @samp{v}
38006 packets then it is possile that @value{GDBN} may run into problems in
38007 other areas, specifically around use of @samp{vFile:setfs:}.
38009 @item vRun;@var{filename}@r{[};@var{argument}@r{]}@dots{}
38010 @cindex @samp{vRun} packet
38011 Run the program @var{filename}, passing it each @var{argument} on its
38012 command line. The file and arguments are hex-encoded strings. If
38013 @var{filename} is an empty string, the stub may use a default program
38014 (e.g.@: the last program run). The program is created in the stopped
38017 @c FIXME: What about non-stop mode?
38019 This packet is only available in extended mode (@pxref{extended mode}).
38025 @item @r{Any stop packet}
38026 for success (@pxref{Stop Reply Packets})
38030 @cindex @samp{vStopped} packet
38031 @xref{Notification Packets}.
38033 @item X @var{addr},@var{length}:@var{XX@dots{}}
38035 @cindex @samp{X} packet
38036 Write data to memory, where the data is transmitted in binary.
38037 Memory is specified by its address @var{addr} and number of addressable memory
38038 units @var{length} (@pxref{addressable memory unit});
38039 @samp{@var{XX}@dots{}} is binary data (@pxref{Binary Data}).
38049 @item z @var{type},@var{addr},@var{kind}
38050 @itemx Z @var{type},@var{addr},@var{kind}
38051 @anchor{insert breakpoint or watchpoint packet}
38052 @cindex @samp{z} packet
38053 @cindex @samp{Z} packets
38054 Insert (@samp{Z}) or remove (@samp{z}) a @var{type} breakpoint or
38055 watchpoint starting at address @var{address} of kind @var{kind}.
38057 Each breakpoint and watchpoint packet @var{type} is documented
38060 @emph{Implementation notes: A remote target shall return an empty string
38061 for an unrecognized breakpoint or watchpoint packet @var{type}. A
38062 remote target shall support either both or neither of a given
38063 @samp{Z@var{type}@dots{}} and @samp{z@var{type}@dots{}} packet pair. To
38064 avoid potential problems with duplicate packets, the operations should
38065 be implemented in an idempotent way.}
38067 @item z0,@var{addr},@var{kind}
38068 @itemx Z0,@var{addr},@var{kind}@r{[};@var{cond_list}@dots{}@r{]}@r{[};cmds:@var{persist},@var{cmd_list}@dots{}@r{]}
38069 @cindex @samp{z0} packet
38070 @cindex @samp{Z0} packet
38071 Insert (@samp{Z0}) or remove (@samp{z0}) a software breakpoint at address
38072 @var{addr} of type @var{kind}.
38074 A software breakpoint is implemented by replacing the instruction at
38075 @var{addr} with a software breakpoint or trap instruction. The
38076 @var{kind} is target-specific and typically indicates the size of the
38077 breakpoint in bytes that should be inserted. E.g., the @sc{arm} and
38078 @sc{mips} can insert either a 2 or 4 byte breakpoint. Some
38079 architectures have additional meanings for @var{kind}
38080 (@pxref{Architecture-Specific Protocol Details}); if no
38081 architecture-specific value is being used, it should be @samp{0}.
38082 @var{kind} is hex-encoded. @var{cond_list} is an optional list of
38083 conditional expressions in bytecode form that should be evaluated on
38084 the target's side. These are the conditions that should be taken into
38085 consideration when deciding if the breakpoint trigger should be
38086 reported back to @value{GDBN}.
38088 See also the @samp{swbreak} stop reason (@pxref{swbreak stop reason})
38089 for how to best report a software breakpoint event to @value{GDBN}.
38091 The @var{cond_list} parameter is comprised of a series of expressions,
38092 concatenated without separators. Each expression has the following form:
38096 @item X @var{len},@var{expr}
38097 @var{len} is the length of the bytecode expression and @var{expr} is the
38098 actual conditional expression in bytecode form.
38102 The optional @var{cmd_list} parameter introduces commands that may be
38103 run on the target, rather than being reported back to @value{GDBN}.
38104 The parameter starts with a numeric flag @var{persist}; if the flag is
38105 nonzero, then the breakpoint may remain active and the commands
38106 continue to be run even when @value{GDBN} disconnects from the target.
38107 Following this flag is a series of expressions concatenated with no
38108 separators. Each expression has the following form:
38112 @item X @var{len},@var{expr}
38113 @var{len} is the length of the bytecode expression and @var{expr} is the
38114 actual commands expression in bytecode form.
38118 @emph{Implementation note: It is possible for a target to copy or move
38119 code that contains software breakpoints (e.g., when implementing
38120 overlays). The behavior of this packet, in the presence of such a
38121 target, is not defined.}
38133 @item z1,@var{addr},@var{kind}
38134 @itemx Z1,@var{addr},@var{kind}@r{[};@var{cond_list}@dots{}@r{]}@r{[};cmds:@var{persist},@var{cmd_list}@dots{}@r{]}
38135 @cindex @samp{z1} packet
38136 @cindex @samp{Z1} packet
38137 Insert (@samp{Z1}) or remove (@samp{z1}) a hardware breakpoint at
38138 address @var{addr}.
38140 A hardware breakpoint is implemented using a mechanism that is not
38141 dependent on being able to modify the target's memory. The
38142 @var{kind}, @var{cond_list}, and @var{cmd_list} arguments have the
38143 same meaning as in @samp{Z0} packets.
38145 @emph{Implementation note: A hardware breakpoint is not affected by code
38158 @item z2,@var{addr},@var{kind}
38159 @itemx Z2,@var{addr},@var{kind}
38160 @cindex @samp{z2} packet
38161 @cindex @samp{Z2} packet
38162 Insert (@samp{Z2}) or remove (@samp{z2}) a write watchpoint at @var{addr}.
38163 The number of bytes to watch is specified by @var{kind}.
38175 @item z3,@var{addr},@var{kind}
38176 @itemx Z3,@var{addr},@var{kind}
38177 @cindex @samp{z3} packet
38178 @cindex @samp{Z3} packet
38179 Insert (@samp{Z3}) or remove (@samp{z3}) a read watchpoint at @var{addr}.
38180 The number of bytes to watch is specified by @var{kind}.
38192 @item z4,@var{addr},@var{kind}
38193 @itemx Z4,@var{addr},@var{kind}
38194 @cindex @samp{z4} packet
38195 @cindex @samp{Z4} packet
38196 Insert (@samp{Z4}) or remove (@samp{z4}) an access watchpoint at @var{addr}.
38197 The number of bytes to watch is specified by @var{kind}.
38211 @node Stop Reply Packets
38212 @section Stop Reply Packets
38213 @cindex stop reply packets
38215 The @samp{C}, @samp{c}, @samp{S}, @samp{s}, @samp{vCont},
38216 @samp{vAttach}, @samp{vRun}, @samp{vStopped}, and @samp{?} packets can
38217 receive any of the below as a reply. Except for @samp{?}
38218 and @samp{vStopped}, that reply is only returned
38219 when the target halts. In the below the exact meaning of @dfn{signal
38220 number} is defined by the header @file{include/gdb/signals.h} in the
38221 @value{GDBN} source code.
38223 In non-stop mode, the server will simply reply @samp{OK} to commands
38224 such as @samp{vCont}; any stop will be the subject of a future
38225 notification. @xref{Remote Non-Stop}.
38227 As in the description of request packets, we include spaces in the
38228 reply templates for clarity; these are not part of the reply packet's
38229 syntax. No @value{GDBN} stop reply packet uses spaces to separate its
38235 The program received signal number @var{AA} (a two-digit hexadecimal
38236 number). This is equivalent to a @samp{T} response with no
38237 @var{n}:@var{r} pairs.
38239 @item T @var{AA} @var{n1}:@var{r1};@var{n2}:@var{r2};@dots{}
38240 @cindex @samp{T} packet reply
38241 The program received signal number @var{AA} (a two-digit hexadecimal
38242 number). This is equivalent to an @samp{S} response, except that the
38243 @samp{@var{n}:@var{r}} pairs can carry values of important registers
38244 and other information directly in the stop reply packet, reducing
38245 round-trip latency. Single-step and breakpoint traps are reported
38246 this way. Each @samp{@var{n}:@var{r}} pair is interpreted as follows:
38250 If @var{n} is a hexadecimal number, it is a register number, and the
38251 corresponding @var{r} gives that register's value. The data @var{r} is a
38252 series of bytes in target byte order, with each byte given by a
38253 two-digit hex number.
38256 If @var{n} is @samp{thread}, then @var{r} is the @var{thread-id} of
38257 the stopped thread, as specified in @ref{thread-id syntax}.
38260 If @var{n} is @samp{core}, then @var{r} is the hexadecimal number of
38261 the core on which the stop event was detected.
38264 If @var{n} is a recognized @dfn{stop reason}, it describes a more
38265 specific event that stopped the target. The currently defined stop
38266 reasons are listed below. The @var{aa} should be @samp{05}, the trap
38267 signal. At most one stop reason should be present.
38270 Otherwise, @value{GDBN} should ignore this @samp{@var{n}:@var{r}} pair
38271 and go on to the next; this allows us to extend the protocol in the
38275 The currently defined stop reasons are:
38281 The packet indicates a watchpoint hit, and @var{r} is the data address, in
38284 @item syscall_entry
38285 @itemx syscall_return
38286 The packet indicates a syscall entry or return, and @var{r} is the
38287 syscall number, in hex.
38289 @cindex shared library events, remote reply
38291 The packet indicates that the loaded libraries have changed.
38292 @value{GDBN} should use @samp{qXfer:libraries:read} to fetch a new
38293 list of loaded libraries. The @var{r} part is ignored.
38295 @cindex replay log events, remote reply
38297 The packet indicates that the target cannot continue replaying
38298 logged execution events, because it has reached the end (or the
38299 beginning when executing backward) of the log. The value of @var{r}
38300 will be either @samp{begin} or @samp{end}. @xref{Reverse Execution},
38301 for more information.
38304 @anchor{swbreak stop reason}
38305 The packet indicates a software breakpoint instruction was executed,
38306 irrespective of whether it was @value{GDBN} that planted the
38307 breakpoint or the breakpoint is hardcoded in the program. The @var{r}
38308 part must be left empty.
38310 On some architectures, such as x86, at the architecture level, when a
38311 breakpoint instruction executes the program counter points at the
38312 breakpoint address plus an offset. On such targets, the stub is
38313 responsible for adjusting the PC to point back at the breakpoint
38316 This packet should not be sent by default; older @value{GDBN} versions
38317 did not support it. @value{GDBN} requests it, by supplying an
38318 appropriate @samp{qSupported} feature (@pxref{qSupported}). The
38319 remote stub must also supply the appropriate @samp{qSupported} feature
38320 indicating support.
38322 This packet is required for correct non-stop mode operation.
38325 The packet indicates the target stopped for a hardware breakpoint.
38326 The @var{r} part must be left empty.
38328 The same remarks about @samp{qSupported} and non-stop mode above
38331 @cindex fork events, remote reply
38333 The packet indicates that @code{fork} was called, and @var{r}
38334 is the thread ID of the new child process. Refer to
38335 @ref{thread-id syntax} for the format of the @var{thread-id}
38336 field. This packet is only applicable to targets that support
38339 This packet should not be sent by default; older @value{GDBN} versions
38340 did not support it. @value{GDBN} requests it, by supplying an
38341 appropriate @samp{qSupported} feature (@pxref{qSupported}). The
38342 remote stub must also supply the appropriate @samp{qSupported} feature
38343 indicating support.
38345 @cindex vfork events, remote reply
38347 The packet indicates that @code{vfork} was called, and @var{r}
38348 is the thread ID of the new child process. Refer to
38349 @ref{thread-id syntax} for the format of the @var{thread-id}
38350 field. This packet is only applicable to targets that support
38353 This packet should not be sent by default; older @value{GDBN} versions
38354 did not support it. @value{GDBN} requests it, by supplying an
38355 appropriate @samp{qSupported} feature (@pxref{qSupported}). The
38356 remote stub must also supply the appropriate @samp{qSupported} feature
38357 indicating support.
38359 @cindex vforkdone events, remote reply
38361 The packet indicates that a child process created by a vfork
38362 has either called @code{exec} or terminated, so that the
38363 address spaces of the parent and child process are no longer
38364 shared. The @var{r} part is ignored. This packet is only
38365 applicable to targets that support vforkdone events.
38367 This packet should not be sent by default; older @value{GDBN} versions
38368 did not support it. @value{GDBN} requests it, by supplying an
38369 appropriate @samp{qSupported} feature (@pxref{qSupported}). The
38370 remote stub must also supply the appropriate @samp{qSupported} feature
38371 indicating support.
38373 @cindex exec events, remote reply
38375 The packet indicates that @code{execve} was called, and @var{r}
38376 is the absolute pathname of the file that was executed, in hex.
38377 This packet is only applicable to targets that support exec events.
38379 This packet should not be sent by default; older @value{GDBN} versions
38380 did not support it. @value{GDBN} requests it, by supplying an
38381 appropriate @samp{qSupported} feature (@pxref{qSupported}). The
38382 remote stub must also supply the appropriate @samp{qSupported} feature
38383 indicating support.
38385 @cindex thread create event, remote reply
38386 @anchor{thread create event}
38388 The packet indicates that the thread was just created. The new thread
38389 is stopped until @value{GDBN} sets it running with a resumption packet
38390 (@pxref{vCont packet}). This packet should not be sent by default;
38391 @value{GDBN} requests it with the @ref{QThreadEvents} packet. See
38392 also the @samp{w} (@pxref{thread exit event}) remote reply below. The
38393 @var{r} part is ignored.
38398 @itemx W @var{AA} ; process:@var{pid}
38399 The process exited, and @var{AA} is the exit status. This is only
38400 applicable to certain targets.
38402 The second form of the response, including the process ID of the
38403 exited process, can be used only when @value{GDBN} has reported
38404 support for multiprocess protocol extensions; see @ref{multiprocess
38405 extensions}. Both @var{AA} and @var{pid} are formatted as big-endian
38409 @itemx X @var{AA} ; process:@var{pid}
38410 The process terminated with signal @var{AA}.
38412 The second form of the response, including the process ID of the
38413 terminated process, can be used only when @value{GDBN} has reported
38414 support for multiprocess protocol extensions; see @ref{multiprocess
38415 extensions}. Both @var{AA} and @var{pid} are formatted as big-endian
38418 @anchor{thread exit event}
38419 @cindex thread exit event, remote reply
38420 @item w @var{AA} ; @var{tid}
38422 The thread exited, and @var{AA} is the exit status. This response
38423 should not be sent by default; @value{GDBN} requests it with the
38424 @ref{QThreadEvents} packet. See also @ref{thread create event} above.
38425 @var{AA} is formatted as a big-endian hex string.
38428 There are no resumed threads left in the target. In other words, even
38429 though the process is alive, the last resumed thread has exited. For
38430 example, say the target process has two threads: thread 1 and thread
38431 2. The client leaves thread 1 stopped, and resumes thread 2, which
38432 subsequently exits. At this point, even though the process is still
38433 alive, and thus no @samp{W} stop reply is sent, no thread is actually
38434 executing either. The @samp{N} stop reply thus informs the client
38435 that it can stop waiting for stop replies. This packet should not be
38436 sent by default; older @value{GDBN} versions did not support it.
38437 @value{GDBN} requests it, by supplying an appropriate
38438 @samp{qSupported} feature (@pxref{qSupported}). The remote stub must
38439 also supply the appropriate @samp{qSupported} feature indicating
38442 @item O @var{XX}@dots{}
38443 @samp{@var{XX}@dots{}} is hex encoding of @sc{ascii} data, to be
38444 written as the program's console output. This can happen at any time
38445 while the program is running and the debugger should continue to wait
38446 for @samp{W}, @samp{T}, etc. This reply is not permitted in non-stop mode.
38448 @item F @var{call-id},@var{parameter}@dots{}
38449 @var{call-id} is the identifier which says which host system call should
38450 be called. This is just the name of the function. Translation into the
38451 correct system call is only applicable as it's defined in @value{GDBN}.
38452 @xref{File-I/O Remote Protocol Extension}, for a list of implemented
38455 @samp{@var{parameter}@dots{}} is a list of parameters as defined for
38456 this very system call.
38458 The target replies with this packet when it expects @value{GDBN} to
38459 call a host system call on behalf of the target. @value{GDBN} replies
38460 with an appropriate @samp{F} packet and keeps up waiting for the next
38461 reply packet from the target. The latest @samp{C}, @samp{c}, @samp{S}
38462 or @samp{s} action is expected to be continued. @xref{File-I/O Remote
38463 Protocol Extension}, for more details.
38467 @node General Query Packets
38468 @section General Query Packets
38469 @cindex remote query requests
38471 Packets starting with @samp{q} are @dfn{general query packets};
38472 packets starting with @samp{Q} are @dfn{general set packets}. General
38473 query and set packets are a semi-unified form for retrieving and
38474 sending information to and from the stub.
38476 The initial letter of a query or set packet is followed by a name
38477 indicating what sort of thing the packet applies to. For example,
38478 @value{GDBN} may use a @samp{qSymbol} packet to exchange symbol
38479 definitions with the stub. These packet names follow some
38484 The name must not contain commas, colons or semicolons.
38486 Most @value{GDBN} query and set packets have a leading upper case
38489 The names of custom vendor packets should use a company prefix, in
38490 lower case, followed by a period. For example, packets designed at
38491 the Acme Corporation might begin with @samp{qacme.foo} (for querying
38492 foos) or @samp{Qacme.bar} (for setting bars).
38495 The name of a query or set packet should be separated from any
38496 parameters by a @samp{:}; the parameters themselves should be
38497 separated by @samp{,} or @samp{;}. Stubs must be careful to match the
38498 full packet name, and check for a separator or the end of the packet,
38499 in case two packet names share a common prefix. New packets should not begin
38500 with @samp{qC}, @samp{qP}, or @samp{qL}@footnote{The @samp{qP} and @samp{qL}
38501 packets predate these conventions, and have arguments without any terminator
38502 for the packet name; we suspect they are in widespread use in places that
38503 are difficult to upgrade. The @samp{qC} packet has no arguments, but some
38504 existing stubs (e.g.@: RedBoot) are known to not check for the end of the
38507 Like the descriptions of the other packets, each description here
38508 has a template showing the packet's overall syntax, followed by an
38509 explanation of the packet's meaning. We include spaces in some of the
38510 templates for clarity; these are not part of the packet's syntax. No
38511 @value{GDBN} packet uses spaces to separate its components.
38513 Here are the currently defined query and set packets:
38519 Turn on or off the agent as a helper to perform some debugging operations
38520 delegated from @value{GDBN} (@pxref{Control Agent}).
38522 @item QAllow:@var{op}:@var{val}@dots{}
38523 @cindex @samp{QAllow} packet
38524 Specify which operations @value{GDBN} expects to request of the
38525 target, as a semicolon-separated list of operation name and value
38526 pairs. Possible values for @var{op} include @samp{WriteReg},
38527 @samp{WriteMem}, @samp{InsertBreak}, @samp{InsertTrace},
38528 @samp{InsertFastTrace}, and @samp{Stop}. @var{val} is either 0,
38529 indicating that @value{GDBN} will not request the operation, or 1,
38530 indicating that it may. (The target can then use this to set up its
38531 own internals optimally, for instance if the debugger never expects to
38532 insert breakpoints, it may not need to install its own trap handler.)
38535 @cindex current thread, remote request
38536 @cindex @samp{qC} packet
38537 Return the current thread ID.
38541 @item QC @var{thread-id}
38542 Where @var{thread-id} is a thread ID as documented in
38543 @ref{thread-id syntax}.
38544 @item @r{(anything else)}
38545 Any other reply implies the old thread ID.
38548 @item qCRC:@var{addr},@var{length}
38549 @cindex CRC of memory block, remote request
38550 @cindex @samp{qCRC} packet
38551 @anchor{qCRC packet}
38552 Compute the CRC checksum of a block of memory using CRC-32 defined in
38553 IEEE 802.3. The CRC is computed byte at a time, taking the most
38554 significant bit of each byte first. The initial pattern code
38555 @code{0xffffffff} is used to ensure leading zeros affect the CRC.
38557 @emph{Note:} This is the same CRC used in validating separate debug
38558 files (@pxref{Separate Debug Files, , Debugging Information in Separate
38559 Files}). However the algorithm is slightly different. When validating
38560 separate debug files, the CRC is computed taking the @emph{least}
38561 significant bit of each byte first, and the final result is inverted to
38562 detect trailing zeros.
38567 An error (such as memory fault)
38568 @item C @var{crc32}
38569 The specified memory region's checksum is @var{crc32}.
38572 @item QDisableRandomization:@var{value}
38573 @cindex disable address space randomization, remote request
38574 @cindex @samp{QDisableRandomization} packet
38575 Some target operating systems will randomize the virtual address space
38576 of the inferior process as a security feature, but provide a feature
38577 to disable such randomization, e.g.@: to allow for a more deterministic
38578 debugging experience. On such systems, this packet with a @var{value}
38579 of 1 directs the target to disable address space randomization for
38580 processes subsequently started via @samp{vRun} packets, while a packet
38581 with a @var{value} of 0 tells the target to enable address space
38584 This packet is only available in extended mode (@pxref{extended mode}).
38589 The request succeeded.
38592 An error occurred. The error number @var{nn} is given as hex digits.
38595 An empty reply indicates that @samp{QDisableRandomization} is not supported
38599 This packet is not probed by default; the remote stub must request it,
38600 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
38601 This should only be done on targets that actually support disabling
38602 address space randomization.
38604 @item QStartupWithShell:@var{value}
38605 @cindex startup with shell, remote request
38606 @cindex @samp{QStartupWithShell} packet
38607 On UNIX-like targets, it is possible to start the inferior using a
38608 shell program. This is the default behavior on both @value{GDBN} and
38609 @command{gdbserver} (@pxref{set startup-with-shell}). This packet is
38610 used to inform @command{gdbserver} whether it should start the
38611 inferior using a shell or not.
38613 If @var{value} is @samp{0}, @command{gdbserver} will not use a shell
38614 to start the inferior. If @var{value} is @samp{1},
38615 @command{gdbserver} will use a shell to start the inferior. All other
38616 values are considered an error.
38618 This packet is only available in extended mode (@pxref{extended
38624 The request succeeded.
38627 An error occurred. The error number @var{nn} is given as hex digits.
38630 This packet is not probed by default; the remote stub must request it,
38631 by supplying an appropriate @samp{qSupported} response
38632 (@pxref{qSupported}). This should only be done on targets that
38633 actually support starting the inferior using a shell.
38635 Use of this packet is controlled by the @code{set startup-with-shell}
38636 command; @pxref{set startup-with-shell}.
38638 @item QEnvironmentHexEncoded:@var{hex-value}
38639 @anchor{QEnvironmentHexEncoded}
38640 @cindex set environment variable, remote request
38641 @cindex @samp{QEnvironmentHexEncoded} packet
38642 On UNIX-like targets, it is possible to set environment variables that
38643 will be passed to the inferior during the startup process. This
38644 packet is used to inform @command{gdbserver} of an environment
38645 variable that has been defined by the user on @value{GDBN} (@pxref{set
38648 The packet is composed by @var{hex-value}, an hex encoded
38649 representation of the @var{name=value} format representing an
38650 environment variable. The name of the environment variable is
38651 represented by @var{name}, and the value to be assigned to the
38652 environment variable is represented by @var{value}. If the variable
38653 has no value (i.e., the value is @code{null}), then @var{value} will
38656 This packet is only available in extended mode (@pxref{extended
38662 The request succeeded.
38665 This packet is not probed by default; the remote stub must request it,
38666 by supplying an appropriate @samp{qSupported} response
38667 (@pxref{qSupported}). This should only be done on targets that
38668 actually support passing environment variables to the starting
38671 This packet is related to the @code{set environment} command;
38672 @pxref{set environment}.
38674 @item QEnvironmentUnset:@var{hex-value}
38675 @anchor{QEnvironmentUnset}
38676 @cindex unset environment variable, remote request
38677 @cindex @samp{QEnvironmentUnset} packet
38678 On UNIX-like targets, it is possible to unset environment variables
38679 before starting the inferior in the remote target. This packet is
38680 used to inform @command{gdbserver} of an environment variable that has
38681 been unset by the user on @value{GDBN} (@pxref{unset environment}).
38683 The packet is composed by @var{hex-value}, an hex encoded
38684 representation of the name of the environment variable to be unset.
38686 This packet is only available in extended mode (@pxref{extended
38692 The request succeeded.
38695 This packet is not probed by default; the remote stub must request it,
38696 by supplying an appropriate @samp{qSupported} response
38697 (@pxref{qSupported}). This should only be done on targets that
38698 actually support passing environment variables to the starting
38701 This packet is related to the @code{unset environment} command;
38702 @pxref{unset environment}.
38704 @item QEnvironmentReset
38705 @anchor{QEnvironmentReset}
38706 @cindex reset environment, remote request
38707 @cindex @samp{QEnvironmentReset} packet
38708 On UNIX-like targets, this packet is used to reset the state of
38709 environment variables in the remote target before starting the
38710 inferior. In this context, reset means unsetting all environment
38711 variables that were previously set by the user (i.e., were not
38712 initially present in the environment). It is sent to
38713 @command{gdbserver} before the @samp{QEnvironmentHexEncoded}
38714 (@pxref{QEnvironmentHexEncoded}) and the @samp{QEnvironmentUnset}
38715 (@pxref{QEnvironmentUnset}) packets.
38717 This packet is only available in extended mode (@pxref{extended
38723 The request succeeded.
38726 This packet is not probed by default; the remote stub must request it,
38727 by supplying an appropriate @samp{qSupported} response
38728 (@pxref{qSupported}). This should only be done on targets that
38729 actually support passing environment variables to the starting
38732 @item QSetWorkingDir:@r{[}@var{directory}@r{]}
38733 @anchor{QSetWorkingDir packet}
38734 @cindex set working directory, remote request
38735 @cindex @samp{QSetWorkingDir} packet
38736 This packet is used to inform the remote server of the intended
38737 current working directory for programs that are going to be executed.
38739 The packet is composed by @var{directory}, an hex encoded
38740 representation of the directory that the remote inferior will use as
38741 its current working directory. If @var{directory} is an empty string,
38742 the remote server should reset the inferior's current working
38743 directory to its original, empty value.
38745 This packet is only available in extended mode (@pxref{extended
38751 The request succeeded.
38755 @itemx qsThreadInfo
38756 @cindex list active threads, remote request
38757 @cindex @samp{qfThreadInfo} packet
38758 @cindex @samp{qsThreadInfo} packet
38759 Obtain a list of all active thread IDs from the target (OS). Since there
38760 may be too many active threads to fit into one reply packet, this query
38761 works iteratively: it may require more than one query/reply sequence to
38762 obtain the entire list of threads. The first query of the sequence will
38763 be the @samp{qfThreadInfo} query; subsequent queries in the
38764 sequence will be the @samp{qsThreadInfo} query.
38766 NOTE: This packet replaces the @samp{qL} query (see below).
38770 @item m @var{thread-id}
38772 @item m @var{thread-id},@var{thread-id}@dots{}
38773 a comma-separated list of thread IDs
38775 (lower case letter @samp{L}) denotes end of list.
38778 In response to each query, the target will reply with a list of one or
38779 more thread IDs, separated by commas.
38780 @value{GDBN} will respond to each reply with a request for more thread
38781 ids (using the @samp{qs} form of the query), until the target responds
38782 with @samp{l} (lower-case ell, for @dfn{last}).
38783 Refer to @ref{thread-id syntax}, for the format of the @var{thread-id}
38786 @emph{Note: @value{GDBN} will send the @code{qfThreadInfo} query during the
38787 initial connection with the remote target, and the very first thread ID
38788 mentioned in the reply will be stopped by @value{GDBN} in a subsequent
38789 message. Therefore, the stub should ensure that the first thread ID in
38790 the @code{qfThreadInfo} reply is suitable for being stopped by @value{GDBN}.}
38792 @item qGetTLSAddr:@var{thread-id},@var{offset},@var{lm}
38793 @cindex get thread-local storage address, remote request
38794 @cindex @samp{qGetTLSAddr} packet
38795 Fetch the address associated with thread local storage specified
38796 by @var{thread-id}, @var{offset}, and @var{lm}.
38798 @var{thread-id} is the thread ID associated with the
38799 thread for which to fetch the TLS address. @xref{thread-id syntax}.
38801 @var{offset} is the (big endian, hex encoded) offset associated with the
38802 thread local variable. (This offset is obtained from the debug
38803 information associated with the variable.)
38805 @var{lm} is the (big endian, hex encoded) OS/ABI-specific encoding of the
38806 load module associated with the thread local storage. For example,
38807 a @sc{gnu}/Linux system will pass the link map address of the shared
38808 object associated with the thread local storage under consideration.
38809 Other operating environments may choose to represent the load module
38810 differently, so the precise meaning of this parameter will vary.
38814 @item @var{XX}@dots{}
38815 Hex encoded (big endian) bytes representing the address of the thread
38816 local storage requested.
38819 An error occurred. The error number @var{nn} is given as hex digits.
38822 An empty reply indicates that @samp{qGetTLSAddr} is not supported by the stub.
38825 @item qGetTIBAddr:@var{thread-id}
38826 @cindex get thread information block address
38827 @cindex @samp{qGetTIBAddr} packet
38828 Fetch address of the Windows OS specific Thread Information Block.
38830 @var{thread-id} is the thread ID associated with the thread.
38834 @item @var{XX}@dots{}
38835 Hex encoded (big endian) bytes representing the linear address of the
38836 thread information block.
38839 An error occured. This means that either the thread was not found, or the
38840 address could not be retrieved.
38843 An empty reply indicates that @samp{qGetTIBAddr} is not supported by the stub.
38846 @item qL @var{startflag} @var{threadcount} @var{nextthread}
38847 Obtain thread information from RTOS. Where: @var{startflag} (one hex
38848 digit) is one to indicate the first query and zero to indicate a
38849 subsequent query; @var{threadcount} (two hex digits) is the maximum
38850 number of threads the response packet can contain; and @var{nextthread}
38851 (eight hex digits), for subsequent queries (@var{startflag} is zero), is
38852 returned in the response as @var{argthread}.
38854 Don't use this packet; use the @samp{qfThreadInfo} query instead (see above).
38858 @item qM @var{count} @var{done} @var{argthread} @var{thread}@dots{}
38859 Where: @var{count} (two hex digits) is the number of threads being
38860 returned; @var{done} (one hex digit) is zero to indicate more threads
38861 and one indicates no further threads; @var{argthreadid} (eight hex
38862 digits) is @var{nextthread} from the request packet; @var{thread}@dots{}
38863 is a sequence of thread IDs, @var{threadid} (eight hex
38864 digits), from the target. See @code{remote.c:parse_threadlist_response()}.
38868 @cindex section offsets, remote request
38869 @cindex @samp{qOffsets} packet
38870 Get section offsets that the target used when relocating the downloaded
38875 @item Text=@var{xxx};Data=@var{yyy}@r{[};Bss=@var{zzz}@r{]}
38876 Relocate the @code{Text} section by @var{xxx} from its original address.
38877 Relocate the @code{Data} section by @var{yyy} from its original address.
38878 If the object file format provides segment information (e.g.@: @sc{elf}
38879 @samp{PT_LOAD} program headers), @value{GDBN} will relocate entire
38880 segments by the supplied offsets.
38882 @emph{Note: while a @code{Bss} offset may be included in the response,
38883 @value{GDBN} ignores this and instead applies the @code{Data} offset
38884 to the @code{Bss} section.}
38886 @item TextSeg=@var{xxx}@r{[};DataSeg=@var{yyy}@r{]}
38887 Relocate the first segment of the object file, which conventionally
38888 contains program code, to a starting address of @var{xxx}. If
38889 @samp{DataSeg} is specified, relocate the second segment, which
38890 conventionally contains modifiable data, to a starting address of
38891 @var{yyy}. @value{GDBN} will report an error if the object file
38892 does not contain segment information, or does not contain at least
38893 as many segments as mentioned in the reply. Extra segments are
38894 kept at fixed offsets relative to the last relocated segment.
38897 @item qP @var{mode} @var{thread-id}
38898 @cindex thread information, remote request
38899 @cindex @samp{qP} packet
38900 Returns information on @var{thread-id}. Where: @var{mode} is a hex
38901 encoded 32 bit mode; @var{thread-id} is a thread ID
38902 (@pxref{thread-id syntax}).
38904 Don't use this packet; use the @samp{qThreadExtraInfo} query instead
38907 Reply: see @code{remote.c:remote_unpack_thread_info_response()}.
38911 @cindex non-stop mode, remote request
38912 @cindex @samp{QNonStop} packet
38914 Enter non-stop (@samp{QNonStop:1}) or all-stop (@samp{QNonStop:0}) mode.
38915 @xref{Remote Non-Stop}, for more information.
38920 The request succeeded.
38923 An error occurred. The error number @var{nn} is given as hex digits.
38926 An empty reply indicates that @samp{QNonStop} is not supported by
38930 This packet is not probed by default; the remote stub must request it,
38931 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
38932 Use of this packet is controlled by the @code{set non-stop} command;
38933 @pxref{Non-Stop Mode}.
38935 @item QCatchSyscalls:1 @r{[};@var{sysno}@r{]}@dots{}
38936 @itemx QCatchSyscalls:0
38937 @cindex catch syscalls from inferior, remote request
38938 @cindex @samp{QCatchSyscalls} packet
38939 @anchor{QCatchSyscalls}
38940 Enable (@samp{QCatchSyscalls:1}) or disable (@samp{QCatchSyscalls:0})
38941 catching syscalls from the inferior process.
38943 For @samp{QCatchSyscalls:1}, each listed syscall @var{sysno} (encoded
38944 in hex) should be reported to @value{GDBN}. If no syscall @var{sysno}
38945 is listed, every system call should be reported.
38947 Note that if a syscall not in the list is reported, @value{GDBN} will
38948 still filter the event according to its own list from all corresponding
38949 @code{catch syscall} commands. However, it is more efficient to only
38950 report the requested syscalls.
38952 Multiple @samp{QCatchSyscalls:1} packets do not combine; any earlier
38953 @samp{QCatchSyscalls:1} list is completely replaced by the new list.
38955 If the inferior process execs, the state of @samp{QCatchSyscalls} is
38956 kept for the new process too. On targets where exec may affect syscall
38957 numbers, for example with exec between 32 and 64-bit processes, the
38958 client should send a new packet with the new syscall list.
38963 The request succeeded.
38966 An error occurred. @var{nn} are hex digits.
38969 An empty reply indicates that @samp{QCatchSyscalls} is not supported by
38973 Use of this packet is controlled by the @code{set remote catch-syscalls}
38974 command (@pxref{Remote Configuration, set remote catch-syscalls}).
38975 This packet is not probed by default; the remote stub must request it,
38976 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
38978 @item QPassSignals: @var{signal} @r{[};@var{signal}@r{]}@dots{}
38979 @cindex pass signals to inferior, remote request
38980 @cindex @samp{QPassSignals} packet
38981 @anchor{QPassSignals}
38982 Each listed @var{signal} should be passed directly to the inferior process.
38983 Signals are numbered identically to continue packets and stop replies
38984 (@pxref{Stop Reply Packets}). Each @var{signal} list item should be
38985 strictly greater than the previous item. These signals do not need to stop
38986 the inferior, or be reported to @value{GDBN}. All other signals should be
38987 reported to @value{GDBN}. Multiple @samp{QPassSignals} packets do not
38988 combine; any earlier @samp{QPassSignals} list is completely replaced by the
38989 new list. This packet improves performance when using @samp{handle
38990 @var{signal} nostop noprint pass}.
38995 The request succeeded.
38998 An error occurred. The error number @var{nn} is given as hex digits.
39001 An empty reply indicates that @samp{QPassSignals} is not supported by
39005 Use of this packet is controlled by the @code{set remote pass-signals}
39006 command (@pxref{Remote Configuration, set remote pass-signals}).
39007 This packet is not probed by default; the remote stub must request it,
39008 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
39010 @item QProgramSignals: @var{signal} @r{[};@var{signal}@r{]}@dots{}
39011 @cindex signals the inferior may see, remote request
39012 @cindex @samp{QProgramSignals} packet
39013 @anchor{QProgramSignals}
39014 Each listed @var{signal} may be delivered to the inferior process.
39015 Others should be silently discarded.
39017 In some cases, the remote stub may need to decide whether to deliver a
39018 signal to the program or not without @value{GDBN} involvement. One
39019 example of that is while detaching --- the program's threads may have
39020 stopped for signals that haven't yet had a chance of being reported to
39021 @value{GDBN}, and so the remote stub can use the signal list specified
39022 by this packet to know whether to deliver or ignore those pending
39025 This does not influence whether to deliver a signal as requested by a
39026 resumption packet (@pxref{vCont packet}).
39028 Signals are numbered identically to continue packets and stop replies
39029 (@pxref{Stop Reply Packets}). Each @var{signal} list item should be
39030 strictly greater than the previous item. Multiple
39031 @samp{QProgramSignals} packets do not combine; any earlier
39032 @samp{QProgramSignals} list is completely replaced by the new list.
39037 The request succeeded.
39040 An error occurred. The error number @var{nn} is given as hex digits.
39043 An empty reply indicates that @samp{QProgramSignals} is not supported
39047 Use of this packet is controlled by the @code{set remote program-signals}
39048 command (@pxref{Remote Configuration, set remote program-signals}).
39049 This packet is not probed by default; the remote stub must request it,
39050 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
39052 @anchor{QThreadEvents}
39053 @item QThreadEvents:1
39054 @itemx QThreadEvents:0
39055 @cindex thread create/exit events, remote request
39056 @cindex @samp{QThreadEvents} packet
39058 Enable (@samp{QThreadEvents:1}) or disable (@samp{QThreadEvents:0})
39059 reporting of thread create and exit events. @xref{thread create
39060 event}, for the reply specifications. For example, this is used in
39061 non-stop mode when @value{GDBN} stops a set of threads and
39062 synchronously waits for the their corresponding stop replies. Without
39063 exit events, if one of the threads exits, @value{GDBN} would hang
39064 forever not knowing that it should no longer expect a stop for that
39065 same thread. @value{GDBN} does not enable this feature unless the
39066 stub reports that it supports it by including @samp{QThreadEvents+} in
39067 its @samp{qSupported} reply.
39072 The request succeeded.
39075 An error occurred. The error number @var{nn} is given as hex digits.
39078 An empty reply indicates that @samp{QThreadEvents} is not supported by
39082 Use of this packet is controlled by the @code{set remote thread-events}
39083 command (@pxref{Remote Configuration, set remote thread-events}).
39085 @item qRcmd,@var{command}
39086 @cindex execute remote command, remote request
39087 @cindex @samp{qRcmd} packet
39088 @var{command} (hex encoded) is passed to the local interpreter for
39089 execution. Invalid commands should be reported using the output
39090 string. Before the final result packet, the target may also respond
39091 with a number of intermediate @samp{O@var{output}} console output
39092 packets. @emph{Implementors should note that providing access to a
39093 stubs's interpreter may have security implications}.
39098 A command response with no output.
39100 A command response with the hex encoded output string @var{OUTPUT}.
39102 Indicate a badly formed request.
39104 An empty reply indicates that @samp{qRcmd} is not recognized.
39107 (Note that the @code{qRcmd} packet's name is separated from the
39108 command by a @samp{,}, not a @samp{:}, contrary to the naming
39109 conventions above. Please don't use this packet as a model for new
39112 @item qSearch:memory:@var{address};@var{length};@var{search-pattern}
39113 @cindex searching memory, in remote debugging
39115 @cindex @samp{qSearch:memory} packet
39117 @cindex @samp{qSearch memory} packet
39118 @anchor{qSearch memory}
39119 Search @var{length} bytes at @var{address} for @var{search-pattern}.
39120 Both @var{address} and @var{length} are encoded in hex;
39121 @var{search-pattern} is a sequence of bytes, also hex encoded.
39126 The pattern was not found.
39128 The pattern was found at @var{address}.
39130 A badly formed request or an error was encountered while searching memory.
39132 An empty reply indicates that @samp{qSearch:memory} is not recognized.
39135 @item QStartNoAckMode
39136 @cindex @samp{QStartNoAckMode} packet
39137 @anchor{QStartNoAckMode}
39138 Request that the remote stub disable the normal @samp{+}/@samp{-}
39139 protocol acknowledgments (@pxref{Packet Acknowledgment}).
39144 The stub has switched to no-acknowledgment mode.
39145 @value{GDBN} acknowledges this reponse,
39146 but neither the stub nor @value{GDBN} shall send or expect further
39147 @samp{+}/@samp{-} acknowledgments in the current connection.
39149 An empty reply indicates that the stub does not support no-acknowledgment mode.
39152 @item qSupported @r{[}:@var{gdbfeature} @r{[};@var{gdbfeature}@r{]}@dots{} @r{]}
39153 @cindex supported packets, remote query
39154 @cindex features of the remote protocol
39155 @cindex @samp{qSupported} packet
39156 @anchor{qSupported}
39157 Tell the remote stub about features supported by @value{GDBN}, and
39158 query the stub for features it supports. This packet allows
39159 @value{GDBN} and the remote stub to take advantage of each others'
39160 features. @samp{qSupported} also consolidates multiple feature probes
39161 at startup, to improve @value{GDBN} performance---a single larger
39162 packet performs better than multiple smaller probe packets on
39163 high-latency links. Some features may enable behavior which must not
39164 be on by default, e.g.@: because it would confuse older clients or
39165 stubs. Other features may describe packets which could be
39166 automatically probed for, but are not. These features must be
39167 reported before @value{GDBN} will use them. This ``default
39168 unsupported'' behavior is not appropriate for all packets, but it
39169 helps to keep the initial connection time under control with new
39170 versions of @value{GDBN} which support increasing numbers of packets.
39174 @item @var{stubfeature} @r{[};@var{stubfeature}@r{]}@dots{}
39175 The stub supports or does not support each returned @var{stubfeature},
39176 depending on the form of each @var{stubfeature} (see below for the
39179 An empty reply indicates that @samp{qSupported} is not recognized,
39180 or that no features needed to be reported to @value{GDBN}.
39183 The allowed forms for each feature (either a @var{gdbfeature} in the
39184 @samp{qSupported} packet, or a @var{stubfeature} in the response)
39188 @item @var{name}=@var{value}
39189 The remote protocol feature @var{name} is supported, and associated
39190 with the specified @var{value}. The format of @var{value} depends
39191 on the feature, but it must not include a semicolon.
39193 The remote protocol feature @var{name} is supported, and does not
39194 need an associated value.
39196 The remote protocol feature @var{name} is not supported.
39198 The remote protocol feature @var{name} may be supported, and
39199 @value{GDBN} should auto-detect support in some other way when it is
39200 needed. This form will not be used for @var{gdbfeature} notifications,
39201 but may be used for @var{stubfeature} responses.
39204 Whenever the stub receives a @samp{qSupported} request, the
39205 supplied set of @value{GDBN} features should override any previous
39206 request. This allows @value{GDBN} to put the stub in a known
39207 state, even if the stub had previously been communicating with
39208 a different version of @value{GDBN}.
39210 The following values of @var{gdbfeature} (for the packet sent by @value{GDBN})
39215 This feature indicates whether @value{GDBN} supports multiprocess
39216 extensions to the remote protocol. @value{GDBN} does not use such
39217 extensions unless the stub also reports that it supports them by
39218 including @samp{multiprocess+} in its @samp{qSupported} reply.
39219 @xref{multiprocess extensions}, for details.
39222 This feature indicates that @value{GDBN} supports the XML target
39223 description. If the stub sees @samp{xmlRegisters=} with target
39224 specific strings separated by a comma, it will report register
39228 This feature indicates whether @value{GDBN} supports the
39229 @samp{qRelocInsn} packet (@pxref{Tracepoint Packets,,Relocate
39230 instruction reply packet}).
39233 This feature indicates whether @value{GDBN} supports the swbreak stop
39234 reason in stop replies. @xref{swbreak stop reason}, for details.
39237 This feature indicates whether @value{GDBN} supports the hwbreak stop
39238 reason in stop replies. @xref{swbreak stop reason}, for details.
39241 This feature indicates whether @value{GDBN} supports fork event
39242 extensions to the remote protocol. @value{GDBN} does not use such
39243 extensions unless the stub also reports that it supports them by
39244 including @samp{fork-events+} in its @samp{qSupported} reply.
39247 This feature indicates whether @value{GDBN} supports vfork event
39248 extensions to the remote protocol. @value{GDBN} does not use such
39249 extensions unless the stub also reports that it supports them by
39250 including @samp{vfork-events+} in its @samp{qSupported} reply.
39253 This feature indicates whether @value{GDBN} supports exec event
39254 extensions to the remote protocol. @value{GDBN} does not use such
39255 extensions unless the stub also reports that it supports them by
39256 including @samp{exec-events+} in its @samp{qSupported} reply.
39258 @item vContSupported
39259 This feature indicates whether @value{GDBN} wants to know the
39260 supported actions in the reply to @samp{vCont?} packet.
39263 Stubs should ignore any unknown values for
39264 @var{gdbfeature}. Any @value{GDBN} which sends a @samp{qSupported}
39265 packet supports receiving packets of unlimited length (earlier
39266 versions of @value{GDBN} may reject overly long responses). Additional values
39267 for @var{gdbfeature} may be defined in the future to let the stub take
39268 advantage of new features in @value{GDBN}, e.g.@: incompatible
39269 improvements in the remote protocol---the @samp{multiprocess} feature is
39270 an example of such a feature. The stub's reply should be independent
39271 of the @var{gdbfeature} entries sent by @value{GDBN}; first @value{GDBN}
39272 describes all the features it supports, and then the stub replies with
39273 all the features it supports.
39275 Similarly, @value{GDBN} will silently ignore unrecognized stub feature
39276 responses, as long as each response uses one of the standard forms.
39278 Some features are flags. A stub which supports a flag feature
39279 should respond with a @samp{+} form response. Other features
39280 require values, and the stub should respond with an @samp{=}
39283 Each feature has a default value, which @value{GDBN} will use if
39284 @samp{qSupported} is not available or if the feature is not mentioned
39285 in the @samp{qSupported} response. The default values are fixed; a
39286 stub is free to omit any feature responses that match the defaults.
39288 Not all features can be probed, but for those which can, the probing
39289 mechanism is useful: in some cases, a stub's internal
39290 architecture may not allow the protocol layer to know some information
39291 about the underlying target in advance. This is especially common in
39292 stubs which may be configured for multiple targets.
39294 These are the currently defined stub features and their properties:
39296 @multitable @columnfractions 0.35 0.2 0.12 0.2
39297 @c NOTE: The first row should be @headitem, but we do not yet require
39298 @c a new enough version of Texinfo (4.7) to use @headitem.
39300 @tab Value Required
39304 @item @samp{PacketSize}
39309 @item @samp{qXfer:auxv:read}
39314 @item @samp{qXfer:btrace:read}
39319 @item @samp{qXfer:btrace-conf:read}
39324 @item @samp{qXfer:exec-file:read}
39329 @item @samp{qXfer:features:read}
39334 @item @samp{qXfer:libraries:read}
39339 @item @samp{qXfer:libraries-svr4:read}
39344 @item @samp{augmented-libraries-svr4-read}
39349 @item @samp{qXfer:memory-map:read}
39354 @item @samp{qXfer:sdata:read}
39359 @item @samp{qXfer:spu:read}
39364 @item @samp{qXfer:spu:write}
39369 @item @samp{qXfer:siginfo:read}
39374 @item @samp{qXfer:siginfo:write}
39379 @item @samp{qXfer:threads:read}
39384 @item @samp{qXfer:traceframe-info:read}
39389 @item @samp{qXfer:uib:read}
39394 @item @samp{qXfer:fdpic:read}
39399 @item @samp{Qbtrace:off}
39404 @item @samp{Qbtrace:bts}
39409 @item @samp{Qbtrace:pt}
39414 @item @samp{Qbtrace-conf:bts:size}
39419 @item @samp{Qbtrace-conf:pt:size}
39424 @item @samp{QNonStop}
39429 @item @samp{QCatchSyscalls}
39434 @item @samp{QPassSignals}
39439 @item @samp{QStartNoAckMode}
39444 @item @samp{multiprocess}
39449 @item @samp{ConditionalBreakpoints}
39454 @item @samp{ConditionalTracepoints}
39459 @item @samp{ReverseContinue}
39464 @item @samp{ReverseStep}
39469 @item @samp{TracepointSource}
39474 @item @samp{QAgent}
39479 @item @samp{QAllow}
39484 @item @samp{QDisableRandomization}
39489 @item @samp{EnableDisableTracepoints}
39494 @item @samp{QTBuffer:size}
39499 @item @samp{tracenz}
39504 @item @samp{BreakpointCommands}
39509 @item @samp{swbreak}
39514 @item @samp{hwbreak}
39519 @item @samp{fork-events}
39524 @item @samp{vfork-events}
39529 @item @samp{exec-events}
39534 @item @samp{QThreadEvents}
39539 @item @samp{no-resumed}
39546 These are the currently defined stub features, in more detail:
39549 @cindex packet size, remote protocol
39550 @item PacketSize=@var{bytes}
39551 The remote stub can accept packets up to at least @var{bytes} in
39552 length. @value{GDBN} will send packets up to this size for bulk
39553 transfers, and will never send larger packets. This is a limit on the
39554 data characters in the packet, including the frame and checksum.
39555 There is no trailing NUL byte in a remote protocol packet; if the stub
39556 stores packets in a NUL-terminated format, it should allow an extra
39557 byte in its buffer for the NUL. If this stub feature is not supported,
39558 @value{GDBN} guesses based on the size of the @samp{g} packet response.
39560 @item qXfer:auxv:read
39561 The remote stub understands the @samp{qXfer:auxv:read} packet
39562 (@pxref{qXfer auxiliary vector read}).
39564 @item qXfer:btrace:read
39565 The remote stub understands the @samp{qXfer:btrace:read}
39566 packet (@pxref{qXfer btrace read}).
39568 @item qXfer:btrace-conf:read
39569 The remote stub understands the @samp{qXfer:btrace-conf:read}
39570 packet (@pxref{qXfer btrace-conf read}).
39572 @item qXfer:exec-file:read
39573 The remote stub understands the @samp{qXfer:exec-file:read} packet
39574 (@pxref{qXfer executable filename read}).
39576 @item qXfer:features:read
39577 The remote stub understands the @samp{qXfer:features:read} packet
39578 (@pxref{qXfer target description read}).
39580 @item qXfer:libraries:read
39581 The remote stub understands the @samp{qXfer:libraries:read} packet
39582 (@pxref{qXfer library list read}).
39584 @item qXfer:libraries-svr4:read
39585 The remote stub understands the @samp{qXfer:libraries-svr4:read} packet
39586 (@pxref{qXfer svr4 library list read}).
39588 @item augmented-libraries-svr4-read
39589 The remote stub understands the augmented form of the
39590 @samp{qXfer:libraries-svr4:read} packet
39591 (@pxref{qXfer svr4 library list read}).
39593 @item qXfer:memory-map:read
39594 The remote stub understands the @samp{qXfer:memory-map:read} packet
39595 (@pxref{qXfer memory map read}).
39597 @item qXfer:sdata:read
39598 The remote stub understands the @samp{qXfer:sdata:read} packet
39599 (@pxref{qXfer sdata read}).
39601 @item qXfer:spu:read
39602 The remote stub understands the @samp{qXfer:spu:read} packet
39603 (@pxref{qXfer spu read}).
39605 @item qXfer:spu:write
39606 The remote stub understands the @samp{qXfer:spu:write} packet
39607 (@pxref{qXfer spu write}).
39609 @item qXfer:siginfo:read
39610 The remote stub understands the @samp{qXfer:siginfo:read} packet
39611 (@pxref{qXfer siginfo read}).
39613 @item qXfer:siginfo:write
39614 The remote stub understands the @samp{qXfer:siginfo:write} packet
39615 (@pxref{qXfer siginfo write}).
39617 @item qXfer:threads:read
39618 The remote stub understands the @samp{qXfer:threads:read} packet
39619 (@pxref{qXfer threads read}).
39621 @item qXfer:traceframe-info:read
39622 The remote stub understands the @samp{qXfer:traceframe-info:read}
39623 packet (@pxref{qXfer traceframe info read}).
39625 @item qXfer:uib:read
39626 The remote stub understands the @samp{qXfer:uib:read}
39627 packet (@pxref{qXfer unwind info block}).
39629 @item qXfer:fdpic:read
39630 The remote stub understands the @samp{qXfer:fdpic:read}
39631 packet (@pxref{qXfer fdpic loadmap read}).
39634 The remote stub understands the @samp{QNonStop} packet
39635 (@pxref{QNonStop}).
39637 @item QCatchSyscalls
39638 The remote stub understands the @samp{QCatchSyscalls} packet
39639 (@pxref{QCatchSyscalls}).
39642 The remote stub understands the @samp{QPassSignals} packet
39643 (@pxref{QPassSignals}).
39645 @item QStartNoAckMode
39646 The remote stub understands the @samp{QStartNoAckMode} packet and
39647 prefers to operate in no-acknowledgment mode. @xref{Packet Acknowledgment}.
39650 @anchor{multiprocess extensions}
39651 @cindex multiprocess extensions, in remote protocol
39652 The remote stub understands the multiprocess extensions to the remote
39653 protocol syntax. The multiprocess extensions affect the syntax of
39654 thread IDs in both packets and replies (@pxref{thread-id syntax}), and
39655 add process IDs to the @samp{D} packet and @samp{W} and @samp{X}
39656 replies. Note that reporting this feature indicates support for the
39657 syntactic extensions only, not that the stub necessarily supports
39658 debugging of more than one process at a time. The stub must not use
39659 multiprocess extensions in packet replies unless @value{GDBN} has also
39660 indicated it supports them in its @samp{qSupported} request.
39662 @item qXfer:osdata:read
39663 The remote stub understands the @samp{qXfer:osdata:read} packet
39664 ((@pxref{qXfer osdata read}).
39666 @item ConditionalBreakpoints
39667 The target accepts and implements evaluation of conditional expressions
39668 defined for breakpoints. The target will only report breakpoint triggers
39669 when such conditions are true (@pxref{Conditions, ,Break Conditions}).
39671 @item ConditionalTracepoints
39672 The remote stub accepts and implements conditional expressions defined
39673 for tracepoints (@pxref{Tracepoint Conditions}).
39675 @item ReverseContinue
39676 The remote stub accepts and implements the reverse continue packet
39680 The remote stub accepts and implements the reverse step packet
39683 @item TracepointSource
39684 The remote stub understands the @samp{QTDPsrc} packet that supplies
39685 the source form of tracepoint definitions.
39688 The remote stub understands the @samp{QAgent} packet.
39691 The remote stub understands the @samp{QAllow} packet.
39693 @item QDisableRandomization
39694 The remote stub understands the @samp{QDisableRandomization} packet.
39696 @item StaticTracepoint
39697 @cindex static tracepoints, in remote protocol
39698 The remote stub supports static tracepoints.
39700 @item InstallInTrace
39701 @anchor{install tracepoint in tracing}
39702 The remote stub supports installing tracepoint in tracing.
39704 @item EnableDisableTracepoints
39705 The remote stub supports the @samp{QTEnable} (@pxref{QTEnable}) and
39706 @samp{QTDisable} (@pxref{QTDisable}) packets that allow tracepoints
39707 to be enabled and disabled while a trace experiment is running.
39709 @item QTBuffer:size
39710 The remote stub supports the @samp{QTBuffer:size} (@pxref{QTBuffer-size})
39711 packet that allows to change the size of the trace buffer.
39714 @cindex string tracing, in remote protocol
39715 The remote stub supports the @samp{tracenz} bytecode for collecting strings.
39716 See @ref{Bytecode Descriptions} for details about the bytecode.
39718 @item BreakpointCommands
39719 @cindex breakpoint commands, in remote protocol
39720 The remote stub supports running a breakpoint's command list itself,
39721 rather than reporting the hit to @value{GDBN}.
39724 The remote stub understands the @samp{Qbtrace:off} packet.
39727 The remote stub understands the @samp{Qbtrace:bts} packet.
39730 The remote stub understands the @samp{Qbtrace:pt} packet.
39732 @item Qbtrace-conf:bts:size
39733 The remote stub understands the @samp{Qbtrace-conf:bts:size} packet.
39735 @item Qbtrace-conf:pt:size
39736 The remote stub understands the @samp{Qbtrace-conf:pt:size} packet.
39739 The remote stub reports the @samp{swbreak} stop reason for memory
39743 The remote stub reports the @samp{hwbreak} stop reason for hardware
39747 The remote stub reports the @samp{fork} stop reason for fork events.
39750 The remote stub reports the @samp{vfork} stop reason for vfork events
39751 and vforkdone events.
39754 The remote stub reports the @samp{exec} stop reason for exec events.
39756 @item vContSupported
39757 The remote stub reports the supported actions in the reply to
39758 @samp{vCont?} packet.
39760 @item QThreadEvents
39761 The remote stub understands the @samp{QThreadEvents} packet.
39764 The remote stub reports the @samp{N} stop reply.
39769 @cindex symbol lookup, remote request
39770 @cindex @samp{qSymbol} packet
39771 Notify the target that @value{GDBN} is prepared to serve symbol lookup
39772 requests. Accept requests from the target for the values of symbols.
39777 The target does not need to look up any (more) symbols.
39778 @item qSymbol:@var{sym_name}
39779 The target requests the value of symbol @var{sym_name} (hex encoded).
39780 @value{GDBN} may provide the value by using the
39781 @samp{qSymbol:@var{sym_value}:@var{sym_name}} message, described
39785 @item qSymbol:@var{sym_value}:@var{sym_name}
39786 Set the value of @var{sym_name} to @var{sym_value}.
39788 @var{sym_name} (hex encoded) is the name of a symbol whose value the
39789 target has previously requested.
39791 @var{sym_value} (hex) is the value for symbol @var{sym_name}. If
39792 @value{GDBN} cannot supply a value for @var{sym_name}, then this field
39798 The target does not need to look up any (more) symbols.
39799 @item qSymbol:@var{sym_name}
39800 The target requests the value of a new symbol @var{sym_name} (hex
39801 encoded). @value{GDBN} will continue to supply the values of symbols
39802 (if available), until the target ceases to request them.
39807 @itemx QTDisconnected
39814 @itemx qTMinFTPILen
39816 @xref{Tracepoint Packets}.
39818 @item qThreadExtraInfo,@var{thread-id}
39819 @cindex thread attributes info, remote request
39820 @cindex @samp{qThreadExtraInfo} packet
39821 Obtain from the target OS a printable string description of thread
39822 attributes for the thread @var{thread-id}; see @ref{thread-id syntax},
39823 for the forms of @var{thread-id}. This
39824 string may contain anything that the target OS thinks is interesting
39825 for @value{GDBN} to tell the user about the thread. The string is
39826 displayed in @value{GDBN}'s @code{info threads} display. Some
39827 examples of possible thread extra info strings are @samp{Runnable}, or
39828 @samp{Blocked on Mutex}.
39832 @item @var{XX}@dots{}
39833 Where @samp{@var{XX}@dots{}} is a hex encoding of @sc{ascii} data,
39834 comprising the printable string containing the extra information about
39835 the thread's attributes.
39838 (Note that the @code{qThreadExtraInfo} packet's name is separated from
39839 the command by a @samp{,}, not a @samp{:}, contrary to the naming
39840 conventions above. Please don't use this packet as a model for new
39859 @xref{Tracepoint Packets}.
39861 @item qXfer:@var{object}:read:@var{annex}:@var{offset},@var{length}
39862 @cindex read special object, remote request
39863 @cindex @samp{qXfer} packet
39864 @anchor{qXfer read}
39865 Read uninterpreted bytes from the target's special data area
39866 identified by the keyword @var{object}. Request @var{length} bytes
39867 starting at @var{offset} bytes into the data. The content and
39868 encoding of @var{annex} is specific to @var{object}; it can supply
39869 additional details about what data to access.
39874 Data @var{data} (@pxref{Binary Data}) has been read from the
39875 target. There may be more data at a higher address (although
39876 it is permitted to return @samp{m} even for the last valid
39877 block of data, as long as at least one byte of data was read).
39878 It is possible for @var{data} to have fewer bytes than the @var{length} in the
39882 Data @var{data} (@pxref{Binary Data}) has been read from the target.
39883 There is no more data to be read. It is possible for @var{data} to
39884 have fewer bytes than the @var{length} in the request.
39887 The @var{offset} in the request is at the end of the data.
39888 There is no more data to be read.
39891 The request was malformed, or @var{annex} was invalid.
39894 The offset was invalid, or there was an error encountered reading the data.
39895 The @var{nn} part is a hex-encoded @code{errno} value.
39898 An empty reply indicates the @var{object} string was not recognized by
39899 the stub, or that the object does not support reading.
39902 Here are the specific requests of this form defined so far. All the
39903 @samp{qXfer:@var{object}:read:@dots{}} requests use the same reply
39904 formats, listed above.
39907 @item qXfer:auxv:read::@var{offset},@var{length}
39908 @anchor{qXfer auxiliary vector read}
39909 Access the target's @dfn{auxiliary vector}. @xref{OS Information,
39910 auxiliary vector}. Note @var{annex} must be empty.
39912 This packet is not probed by default; the remote stub must request it,
39913 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
39915 @item qXfer:btrace:read:@var{annex}:@var{offset},@var{length}
39916 @anchor{qXfer btrace read}
39918 Return a description of the current branch trace.
39919 @xref{Branch Trace Format}. The annex part of the generic @samp{qXfer}
39920 packet may have one of the following values:
39924 Returns all available branch trace.
39927 Returns all available branch trace if the branch trace changed since
39928 the last read request.
39931 Returns the new branch trace since the last read request. Adds a new
39932 block to the end of the trace that begins at zero and ends at the source
39933 location of the first branch in the trace buffer. This extra block is
39934 used to stitch traces together.
39936 If the trace buffer overflowed, returns an error indicating the overflow.
39939 This packet is not probed by default; the remote stub must request it
39940 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
39942 @item qXfer:btrace-conf:read::@var{offset},@var{length}
39943 @anchor{qXfer btrace-conf read}
39945 Return a description of the current branch trace configuration.
39946 @xref{Branch Trace Configuration Format}.
39948 This packet is not probed by default; the remote stub must request it
39949 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
39951 @item qXfer:exec-file:read:@var{annex}:@var{offset},@var{length}
39952 @anchor{qXfer executable filename read}
39953 Return the full absolute name of the file that was executed to create
39954 a process running on the remote system. The annex specifies the
39955 numeric process ID of the process to query, encoded as a hexadecimal
39956 number. If the annex part is empty the remote stub should return the
39957 filename corresponding to the currently executing process.
39959 This packet is not probed by default; the remote stub must request it,
39960 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
39962 @item qXfer:features:read:@var{annex}:@var{offset},@var{length}
39963 @anchor{qXfer target description read}
39964 Access the @dfn{target description}. @xref{Target Descriptions}. The
39965 annex specifies which XML document to access. The main description is
39966 always loaded from the @samp{target.xml} annex.
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:libraries:read:@var{annex}:@var{offset},@var{length}
39972 @anchor{qXfer library list read}
39973 Access the target's list of loaded libraries. @xref{Library List Format}.
39974 The annex part of the generic @samp{qXfer} packet must be empty
39975 (@pxref{qXfer read}).
39977 Targets which maintain a list of libraries in the program's memory do
39978 not need to implement this packet; it is designed for platforms where
39979 the operating system manages the list of loaded libraries.
39981 This packet is not probed by default; the remote stub must request it,
39982 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
39984 @item qXfer:libraries-svr4:read:@var{annex}:@var{offset},@var{length}
39985 @anchor{qXfer svr4 library list read}
39986 Access the target's list of loaded libraries when the target is an SVR4
39987 platform. @xref{Library List Format for SVR4 Targets}. The annex part
39988 of the generic @samp{qXfer} packet must be empty unless the remote
39989 stub indicated it supports the augmented form of this packet
39990 by supplying an appropriate @samp{qSupported} response
39991 (@pxref{qXfer read}, @ref{qSupported}).
39993 This packet is optional for better performance on SVR4 targets.
39994 @value{GDBN} uses memory read packets to read the SVR4 library list otherwise.
39996 This packet is not probed by default; the remote stub must request it,
39997 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
39999 If the remote stub indicates it supports the augmented form of this
40000 packet then the annex part of the generic @samp{qXfer} packet may
40001 contain a semicolon-separated list of @samp{@var{name}=@var{value}}
40002 arguments. The currently supported arguments are:
40005 @item start=@var{address}
40006 A hexadecimal number specifying the address of the @samp{struct
40007 link_map} to start reading the library list from. If unset or zero
40008 then the first @samp{struct link_map} in the library list will be
40009 chosen as the starting point.
40011 @item prev=@var{address}
40012 A hexadecimal number specifying the address of the @samp{struct
40013 link_map} immediately preceding the @samp{struct link_map}
40014 specified by the @samp{start} argument. If unset or zero then
40015 the remote stub will expect that no @samp{struct link_map}
40016 exists prior to the starting point.
40020 Arguments that are not understood by the remote stub will be silently
40023 @item qXfer:memory-map:read::@var{offset},@var{length}
40024 @anchor{qXfer memory map read}
40025 Access the target's @dfn{memory-map}. @xref{Memory Map Format}. The
40026 annex part of the generic @samp{qXfer} packet must be empty
40027 (@pxref{qXfer read}).
40029 This packet is not probed by default; the remote stub must request it,
40030 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
40032 @item qXfer:sdata:read::@var{offset},@var{length}
40033 @anchor{qXfer sdata read}
40035 Read contents of the extra collected static tracepoint marker
40036 information. The annex part of the generic @samp{qXfer} packet must
40037 be empty (@pxref{qXfer read}). @xref{Tracepoint Actions,,Tracepoint
40040 This packet is not probed by default; the remote stub must request it,
40041 by supplying an appropriate @samp{qSupported} response
40042 (@pxref{qSupported}).
40044 @item qXfer:siginfo:read::@var{offset},@var{length}
40045 @anchor{qXfer siginfo read}
40046 Read contents of the extra signal information on the target
40047 system. The annex part of the generic @samp{qXfer} packet must be
40048 empty (@pxref{qXfer read}).
40050 This packet is not probed by default; the remote stub must request it,
40051 by supplying an appropriate @samp{qSupported} response
40052 (@pxref{qSupported}).
40054 @item qXfer:spu:read:@var{annex}:@var{offset},@var{length}
40055 @anchor{qXfer spu read}
40056 Read contents of an @code{spufs} file on the target system. The
40057 annex specifies which file to read; it must be of the form
40058 @file{@var{id}/@var{name}}, where @var{id} specifies an SPU context ID
40059 in the target process, and @var{name} identifes the @code{spufs} file
40060 in that context to be accessed.
40062 This packet is not probed by default; the remote stub must request it,
40063 by supplying an appropriate @samp{qSupported} response
40064 (@pxref{qSupported}).
40066 @item qXfer:threads:read::@var{offset},@var{length}
40067 @anchor{qXfer threads read}
40068 Access the list of threads on target. @xref{Thread List Format}. The
40069 annex part of the generic @samp{qXfer} packet must be empty
40070 (@pxref{qXfer read}).
40072 This packet is not probed by default; the remote stub must request it,
40073 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
40075 @item qXfer:traceframe-info:read::@var{offset},@var{length}
40076 @anchor{qXfer traceframe info read}
40078 Return a description of the current traceframe's contents.
40079 @xref{Traceframe Info Format}. The annex part of the generic
40080 @samp{qXfer} packet must be empty (@pxref{qXfer read}).
40082 This packet is not probed by default; the remote stub must request it,
40083 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
40085 @item qXfer:uib:read:@var{pc}:@var{offset},@var{length}
40086 @anchor{qXfer unwind info block}
40088 Return the unwind information block for @var{pc}. This packet is used
40089 on OpenVMS/ia64 to ask the kernel unwind information.
40091 This packet is not probed by default.
40093 @item qXfer:fdpic:read:@var{annex}:@var{offset},@var{length}
40094 @anchor{qXfer fdpic loadmap read}
40095 Read contents of @code{loadmap}s on the target system. The
40096 annex, either @samp{exec} or @samp{interp}, specifies which @code{loadmap},
40097 executable @code{loadmap} or interpreter @code{loadmap} to read.
40099 This packet is not probed by default; the remote stub must request it,
40100 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
40102 @item qXfer:osdata:read::@var{offset},@var{length}
40103 @anchor{qXfer osdata read}
40104 Access the target's @dfn{operating system information}.
40105 @xref{Operating System Information}.
40109 @item qXfer:@var{object}:write:@var{annex}:@var{offset}:@var{data}@dots{}
40110 @cindex write data into object, remote request
40111 @anchor{qXfer write}
40112 Write uninterpreted bytes into the target's special data area
40113 identified by the keyword @var{object}, starting at @var{offset} bytes
40114 into the data. The binary-encoded data (@pxref{Binary Data}) to be
40115 written is given by @var{data}@dots{}. The content and encoding of @var{annex}
40116 is specific to @var{object}; it can supply additional details about what data
40122 @var{nn} (hex encoded) is the number of bytes written.
40123 This may be fewer bytes than supplied in the request.
40126 The request was malformed, or @var{annex} was invalid.
40129 The offset was invalid, or there was an error encountered writing the data.
40130 The @var{nn} part is a hex-encoded @code{errno} value.
40133 An empty reply indicates the @var{object} string was not
40134 recognized by the stub, or that the object does not support writing.
40137 Here are the specific requests of this form defined so far. All the
40138 @samp{qXfer:@var{object}:write:@dots{}} requests use the same reply
40139 formats, listed above.
40142 @item qXfer:siginfo:write::@var{offset}:@var{data}@dots{}
40143 @anchor{qXfer siginfo write}
40144 Write @var{data} to the extra signal information on the target system.
40145 The annex part of the generic @samp{qXfer} packet must be
40146 empty (@pxref{qXfer write}).
40148 This packet is not probed by default; the remote stub must request it,
40149 by supplying an appropriate @samp{qSupported} response
40150 (@pxref{qSupported}).
40152 @item qXfer:spu:write:@var{annex}:@var{offset}:@var{data}@dots{}
40153 @anchor{qXfer spu write}
40154 Write @var{data} to an @code{spufs} file on the target system. The
40155 annex specifies which file to write; it must be of the form
40156 @file{@var{id}/@var{name}}, where @var{id} specifies an SPU context ID
40157 in the target process, and @var{name} identifes the @code{spufs} file
40158 in that context to be accessed.
40160 This packet is not probed by default; the remote stub must request it,
40161 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
40164 @item qXfer:@var{object}:@var{operation}:@dots{}
40165 Requests of this form may be added in the future. When a stub does
40166 not recognize the @var{object} keyword, or its support for
40167 @var{object} does not recognize the @var{operation} keyword, the stub
40168 must respond with an empty packet.
40170 @item qAttached:@var{pid}
40171 @cindex query attached, remote request
40172 @cindex @samp{qAttached} packet
40173 Return an indication of whether the remote server attached to an
40174 existing process or created a new process. When the multiprocess
40175 protocol extensions are supported (@pxref{multiprocess extensions}),
40176 @var{pid} is an integer in hexadecimal format identifying the target
40177 process. Otherwise, @value{GDBN} will omit the @var{pid} field and
40178 the query packet will be simplified as @samp{qAttached}.
40180 This query is used, for example, to know whether the remote process
40181 should be detached or killed when a @value{GDBN} session is ended with
40182 the @code{quit} command.
40187 The remote server attached to an existing process.
40189 The remote server created a new process.
40191 A badly formed request or an error was encountered.
40195 Enable branch tracing for the current thread using Branch Trace Store.
40200 Branch tracing has been enabled.
40202 A badly formed request or an error was encountered.
40206 Enable branch tracing for the current thread using Intel Processor Trace.
40211 Branch tracing has been enabled.
40213 A badly formed request or an error was encountered.
40217 Disable branch tracing for the current thread.
40222 Branch tracing has been disabled.
40224 A badly formed request or an error was encountered.
40227 @item Qbtrace-conf:bts:size=@var{value}
40228 Set the requested ring buffer size for new threads that use the
40229 btrace recording method in bts format.
40234 The ring buffer size has been set.
40236 A badly formed request or an error was encountered.
40239 @item Qbtrace-conf:pt:size=@var{value}
40240 Set the requested ring buffer size for new threads that use the
40241 btrace recording method in pt format.
40246 The ring buffer size has been set.
40248 A badly formed request or an error was encountered.
40253 @node Architecture-Specific Protocol Details
40254 @section Architecture-Specific Protocol Details
40256 This section describes how the remote protocol is applied to specific
40257 target architectures. Also see @ref{Standard Target Features}, for
40258 details of XML target descriptions for each architecture.
40261 * ARM-Specific Protocol Details::
40262 * MIPS-Specific Protocol Details::
40265 @node ARM-Specific Protocol Details
40266 @subsection @acronym{ARM}-specific Protocol Details
40269 * ARM Breakpoint Kinds::
40272 @node ARM Breakpoint Kinds
40273 @subsubsection @acronym{ARM} Breakpoint Kinds
40274 @cindex breakpoint kinds, @acronym{ARM}
40276 These breakpoint kinds are defined for the @samp{Z0} and @samp{Z1} packets.
40281 16-bit Thumb mode breakpoint.
40284 32-bit Thumb mode (Thumb-2) breakpoint.
40287 32-bit @acronym{ARM} mode breakpoint.
40291 @node MIPS-Specific Protocol Details
40292 @subsection @acronym{MIPS}-specific Protocol Details
40295 * MIPS Register packet Format::
40296 * MIPS Breakpoint Kinds::
40299 @node MIPS Register packet Format
40300 @subsubsection @acronym{MIPS} Register Packet Format
40301 @cindex register packet format, @acronym{MIPS}
40303 The following @code{g}/@code{G} packets have previously been defined.
40304 In the below, some thirty-two bit registers are transferred as
40305 sixty-four bits. Those registers should be zero/sign extended (which?)
40306 to fill the space allocated. Register bytes are transferred in target
40307 byte order. The two nibbles within a register byte are transferred
40308 most-significant -- least-significant.
40313 All registers are transferred as thirty-two bit quantities in the order:
40314 32 general-purpose; sr; lo; hi; bad; cause; pc; 32 floating-point
40315 registers; fsr; fir; fp.
40318 All registers are transferred as sixty-four bit quantities (including
40319 thirty-two bit registers such as @code{sr}). The ordering is the same
40324 @node MIPS Breakpoint Kinds
40325 @subsubsection @acronym{MIPS} Breakpoint Kinds
40326 @cindex breakpoint kinds, @acronym{MIPS}
40328 These breakpoint kinds are defined for the @samp{Z0} and @samp{Z1} packets.
40333 16-bit @acronym{MIPS16} mode breakpoint.
40336 16-bit @acronym{microMIPS} mode breakpoint.
40339 32-bit standard @acronym{MIPS} mode breakpoint.
40342 32-bit @acronym{microMIPS} mode breakpoint.
40346 @node Tracepoint Packets
40347 @section Tracepoint Packets
40348 @cindex tracepoint packets
40349 @cindex packets, tracepoint
40351 Here we describe the packets @value{GDBN} uses to implement
40352 tracepoints (@pxref{Tracepoints}).
40356 @item QTDP:@var{n}:@var{addr}:@var{ena}:@var{step}:@var{pass}[:F@var{flen}][:X@var{len},@var{bytes}]@r{[}-@r{]}
40357 @cindex @samp{QTDP} packet
40358 Create a new tracepoint, number @var{n}, at @var{addr}. If @var{ena}
40359 is @samp{E}, then the tracepoint is enabled; if it is @samp{D}, then
40360 the tracepoint is disabled. The @var{step} gives the tracepoint's step
40361 count, and @var{pass} gives its pass count. If an @samp{F} is present,
40362 then the tracepoint is to be a fast tracepoint, and the @var{flen} is
40363 the number of bytes that the target should copy elsewhere to make room
40364 for the tracepoint. If an @samp{X} is present, it introduces a
40365 tracepoint condition, which consists of a hexadecimal length, followed
40366 by a comma and hex-encoded bytes, in a manner similar to action
40367 encodings as described below. If the trailing @samp{-} is present,
40368 further @samp{QTDP} packets will follow to specify this tracepoint's
40374 The packet was understood and carried out.
40376 @xref{Tracepoint Packets,,Relocate instruction reply packet}.
40378 The packet was not recognized.
40381 @item QTDP:-@var{n}:@var{addr}:@r{[}S@r{]}@var{action}@dots{}@r{[}-@r{]}
40382 Define actions to be taken when a tracepoint is hit. The @var{n} and
40383 @var{addr} must be the same as in the initial @samp{QTDP} packet for
40384 this tracepoint. This packet may only be sent immediately after
40385 another @samp{QTDP} packet that ended with a @samp{-}. If the
40386 trailing @samp{-} is present, further @samp{QTDP} packets will follow,
40387 specifying more actions for this tracepoint.
40389 In the series of action packets for a given tracepoint, at most one
40390 can have an @samp{S} before its first @var{action}. If such a packet
40391 is sent, it and the following packets define ``while-stepping''
40392 actions. Any prior packets define ordinary actions --- that is, those
40393 taken when the tracepoint is first hit. If no action packet has an
40394 @samp{S}, then all the packets in the series specify ordinary
40395 tracepoint actions.
40397 The @samp{@var{action}@dots{}} portion of the packet is a series of
40398 actions, concatenated without separators. Each action has one of the
40404 Collect the registers whose bits are set in @var{mask},
40405 a hexadecimal number whose @var{i}'th bit is set if register number
40406 @var{i} should be collected. (The least significant bit is numbered
40407 zero.) Note that @var{mask} may be any number of digits long; it may
40408 not fit in a 32-bit word.
40410 @item M @var{basereg},@var{offset},@var{len}
40411 Collect @var{len} bytes of memory starting at the address in register
40412 number @var{basereg}, plus @var{offset}. If @var{basereg} is
40413 @samp{-1}, then the range has a fixed address: @var{offset} is the
40414 address of the lowest byte to collect. The @var{basereg},
40415 @var{offset}, and @var{len} parameters are all unsigned hexadecimal
40416 values (the @samp{-1} value for @var{basereg} is a special case).
40418 @item X @var{len},@var{expr}
40419 Evaluate @var{expr}, whose length is @var{len}, and collect memory as
40420 it directs. The agent expression @var{expr} is as described in
40421 @ref{Agent Expressions}. Each byte of the expression is encoded as a
40422 two-digit hex number in the packet; @var{len} is the number of bytes
40423 in the expression (and thus one-half the number of hex digits in the
40428 Any number of actions may be packed together in a single @samp{QTDP}
40429 packet, as long as the packet does not exceed the maximum packet
40430 length (400 bytes, for many stubs). There may be only one @samp{R}
40431 action per tracepoint, and it must precede any @samp{M} or @samp{X}
40432 actions. Any registers referred to by @samp{M} and @samp{X} actions
40433 must be collected by a preceding @samp{R} action. (The
40434 ``while-stepping'' actions are treated as if they were attached to a
40435 separate tracepoint, as far as these restrictions are concerned.)
40440 The packet was understood and carried out.
40442 @xref{Tracepoint Packets,,Relocate instruction reply packet}.
40444 The packet was not recognized.
40447 @item QTDPsrc:@var{n}:@var{addr}:@var{type}:@var{start}:@var{slen}:@var{bytes}
40448 @cindex @samp{QTDPsrc} packet
40449 Specify a source string of tracepoint @var{n} at address @var{addr}.
40450 This is useful to get accurate reproduction of the tracepoints
40451 originally downloaded at the beginning of the trace run. The @var{type}
40452 is the name of the tracepoint part, such as @samp{cond} for the
40453 tracepoint's conditional expression (see below for a list of types), while
40454 @var{bytes} is the string, encoded in hexadecimal.
40456 @var{start} is the offset of the @var{bytes} within the overall source
40457 string, while @var{slen} is the total length of the source string.
40458 This is intended for handling source strings that are longer than will
40459 fit in a single packet.
40460 @c Add detailed example when this info is moved into a dedicated
40461 @c tracepoint descriptions section.
40463 The available string types are @samp{at} for the location,
40464 @samp{cond} for the conditional, and @samp{cmd} for an action command.
40465 @value{GDBN} sends a separate packet for each command in the action
40466 list, in the same order in which the commands are stored in the list.
40468 The target does not need to do anything with source strings except
40469 report them back as part of the replies to the @samp{qTfP}/@samp{qTsP}
40472 Although this packet is optional, and @value{GDBN} will only send it
40473 if the target replies with @samp{TracepointSource} @xref{General
40474 Query Packets}, it makes both disconnected tracing and trace files
40475 much easier to use. Otherwise the user must be careful that the
40476 tracepoints in effect while looking at trace frames are identical to
40477 the ones in effect during the trace run; even a small discrepancy
40478 could cause @samp{tdump} not to work, or a particular trace frame not
40481 @item QTDV:@var{n}:@var{value}:@var{builtin}:@var{name}
40482 @cindex define trace state variable, remote request
40483 @cindex @samp{QTDV} packet
40484 Create a new trace state variable, number @var{n}, with an initial
40485 value of @var{value}, which is a 64-bit signed integer. Both @var{n}
40486 and @var{value} are encoded as hexadecimal values. @value{GDBN} has
40487 the option of not using this packet for initial values of zero; the
40488 target should simply create the trace state variables as they are
40489 mentioned in expressions. The value @var{builtin} should be 1 (one)
40490 if the trace state variable is builtin and 0 (zero) if it is not builtin.
40491 @value{GDBN} only sets @var{builtin} to 1 if a previous @samp{qTfV} or
40492 @samp{qTsV} packet had it set. The contents of @var{name} is the
40493 hex-encoded name (without the leading @samp{$}) of the trace state
40496 @item QTFrame:@var{n}
40497 @cindex @samp{QTFrame} packet
40498 Select the @var{n}'th tracepoint frame from the buffer, and use the
40499 register and memory contents recorded there to answer subsequent
40500 request packets from @value{GDBN}.
40502 A successful reply from the stub indicates that the stub has found the
40503 requested frame. The response is a series of parts, concatenated
40504 without separators, describing the frame we selected. Each part has
40505 one of the following forms:
40509 The selected frame is number @var{n} in the trace frame buffer;
40510 @var{f} is a hexadecimal number. If @var{f} is @samp{-1}, then there
40511 was no frame matching the criteria in the request packet.
40514 The selected trace frame records a hit of tracepoint number @var{t};
40515 @var{t} is a hexadecimal number.
40519 @item QTFrame:pc:@var{addr}
40520 Like @samp{QTFrame:@var{n}}, but select the first tracepoint frame after the
40521 currently selected frame whose PC is @var{addr};
40522 @var{addr} is a hexadecimal number.
40524 @item QTFrame:tdp:@var{t}
40525 Like @samp{QTFrame:@var{n}}, but select the first tracepoint frame after the
40526 currently selected frame that is a hit of tracepoint @var{t}; @var{t}
40527 is a hexadecimal number.
40529 @item QTFrame:range:@var{start}:@var{end}
40530 Like @samp{QTFrame:@var{n}}, but select the first tracepoint frame after the
40531 currently selected frame whose PC is between @var{start} (inclusive)
40532 and @var{end} (inclusive); @var{start} and @var{end} are hexadecimal
40535 @item QTFrame:outside:@var{start}:@var{end}
40536 Like @samp{QTFrame:range:@var{start}:@var{end}}, but select the first
40537 frame @emph{outside} the given range of addresses (exclusive).
40540 @cindex @samp{qTMinFTPILen} packet
40541 This packet requests the minimum length of instruction at which a fast
40542 tracepoint (@pxref{Set Tracepoints}) may be placed. For instance, on
40543 the 32-bit x86 architecture, it is possible to use a 4-byte jump, but
40544 it depends on the target system being able to create trampolines in
40545 the first 64K of memory, which might or might not be possible for that
40546 system. So the reply to this packet will be 4 if it is able to
40553 The minimum instruction length is currently unknown.
40555 The minimum instruction length is @var{length}, where @var{length}
40556 is a hexadecimal number greater or equal to 1. A reply
40557 of 1 means that a fast tracepoint may be placed on any instruction
40558 regardless of size.
40560 An error has occurred.
40562 An empty reply indicates that the request is not supported by the stub.
40566 @cindex @samp{QTStart} packet
40567 Begin the tracepoint experiment. Begin collecting data from
40568 tracepoint hits in the trace frame buffer. This packet supports the
40569 @samp{qRelocInsn} reply (@pxref{Tracepoint Packets,,Relocate
40570 instruction reply packet}).
40573 @cindex @samp{QTStop} packet
40574 End the tracepoint experiment. Stop collecting trace frames.
40576 @item QTEnable:@var{n}:@var{addr}
40578 @cindex @samp{QTEnable} packet
40579 Enable tracepoint @var{n} at address @var{addr} in a started tracepoint
40580 experiment. If the tracepoint was previously disabled, then collection
40581 of data from it will resume.
40583 @item QTDisable:@var{n}:@var{addr}
40585 @cindex @samp{QTDisable} packet
40586 Disable tracepoint @var{n} at address @var{addr} in a started tracepoint
40587 experiment. No more data will be collected from the tracepoint unless
40588 @samp{QTEnable:@var{n}:@var{addr}} is subsequently issued.
40591 @cindex @samp{QTinit} packet
40592 Clear the table of tracepoints, and empty the trace frame buffer.
40594 @item QTro:@var{start1},@var{end1}:@var{start2},@var{end2}:@dots{}
40595 @cindex @samp{QTro} packet
40596 Establish the given ranges of memory as ``transparent''. The stub
40597 will answer requests for these ranges from memory's current contents,
40598 if they were not collected as part of the tracepoint hit.
40600 @value{GDBN} uses this to mark read-only regions of memory, like those
40601 containing program code. Since these areas never change, they should
40602 still have the same contents they did when the tracepoint was hit, so
40603 there's no reason for the stub to refuse to provide their contents.
40605 @item QTDisconnected:@var{value}
40606 @cindex @samp{QTDisconnected} packet
40607 Set the choice to what to do with the tracing run when @value{GDBN}
40608 disconnects from the target. A @var{value} of 1 directs the target to
40609 continue the tracing run, while 0 tells the target to stop tracing if
40610 @value{GDBN} is no longer in the picture.
40613 @cindex @samp{qTStatus} packet
40614 Ask the stub if there is a trace experiment running right now.
40616 The reply has the form:
40620 @item T@var{running}@r{[};@var{field}@r{]}@dots{}
40621 @var{running} is a single digit @code{1} if the trace is presently
40622 running, or @code{0} if not. It is followed by semicolon-separated
40623 optional fields that an agent may use to report additional status.
40627 If the trace is not running, the agent may report any of several
40628 explanations as one of the optional fields:
40633 No trace has been run yet.
40635 @item tstop[:@var{text}]:0
40636 The trace was stopped by a user-originated stop command. The optional
40637 @var{text} field is a user-supplied string supplied as part of the
40638 stop command (for instance, an explanation of why the trace was
40639 stopped manually). It is hex-encoded.
40642 The trace stopped because the trace buffer filled up.
40644 @item tdisconnected:0
40645 The trace stopped because @value{GDBN} disconnected from the target.
40647 @item tpasscount:@var{tpnum}
40648 The trace stopped because tracepoint @var{tpnum} exceeded its pass count.
40650 @item terror:@var{text}:@var{tpnum}
40651 The trace stopped because tracepoint @var{tpnum} had an error. The
40652 string @var{text} is available to describe the nature of the error
40653 (for instance, a divide by zero in the condition expression); it
40657 The trace stopped for some other reason.
40661 Additional optional fields supply statistical and other information.
40662 Although not required, they are extremely useful for users monitoring
40663 the progress of a trace run. If a trace has stopped, and these
40664 numbers are reported, they must reflect the state of the just-stopped
40669 @item tframes:@var{n}
40670 The number of trace frames in the buffer.
40672 @item tcreated:@var{n}
40673 The total number of trace frames created during the run. This may
40674 be larger than the trace frame count, if the buffer is circular.
40676 @item tsize:@var{n}
40677 The total size of the trace buffer, in bytes.
40679 @item tfree:@var{n}
40680 The number of bytes still unused in the buffer.
40682 @item circular:@var{n}
40683 The value of the circular trace buffer flag. @code{1} means that the
40684 trace buffer is circular and old trace frames will be discarded if
40685 necessary to make room, @code{0} means that the trace buffer is linear
40688 @item disconn:@var{n}
40689 The value of the disconnected tracing flag. @code{1} means that
40690 tracing will continue after @value{GDBN} disconnects, @code{0} means
40691 that the trace run will stop.
40695 @item qTP:@var{tp}:@var{addr}
40696 @cindex tracepoint status, remote request
40697 @cindex @samp{qTP} packet
40698 Ask the stub for the current state of tracepoint number @var{tp} at
40699 address @var{addr}.
40703 @item V@var{hits}:@var{usage}
40704 The tracepoint has been hit @var{hits} times so far during the trace
40705 run, and accounts for @var{usage} in the trace buffer. Note that
40706 @code{while-stepping} steps are not counted as separate hits, but the
40707 steps' space consumption is added into the usage number.
40711 @item qTV:@var{var}
40712 @cindex trace state variable value, remote request
40713 @cindex @samp{qTV} packet
40714 Ask the stub for the value of the trace state variable number @var{var}.
40719 The value of the variable is @var{value}. This will be the current
40720 value of the variable if the user is examining a running target, or a
40721 saved value if the variable was collected in the trace frame that the
40722 user is looking at. Note that multiple requests may result in
40723 different reply values, such as when requesting values while the
40724 program is running.
40727 The value of the variable is unknown. This would occur, for example,
40728 if the user is examining a trace frame in which the requested variable
40733 @cindex @samp{qTfP} packet
40735 @cindex @samp{qTsP} packet
40736 These packets request data about tracepoints that are being used by
40737 the target. @value{GDBN} sends @code{qTfP} to get the first piece
40738 of data, and multiple @code{qTsP} to get additional pieces. Replies
40739 to these packets generally take the form of the @code{QTDP} packets
40740 that define tracepoints. (FIXME add detailed syntax)
40743 @cindex @samp{qTfV} packet
40745 @cindex @samp{qTsV} packet
40746 These packets request data about trace state variables that are on the
40747 target. @value{GDBN} sends @code{qTfV} to get the first vari of data,
40748 and multiple @code{qTsV} to get additional variables. Replies to
40749 these packets follow the syntax of the @code{QTDV} packets that define
40750 trace state variables.
40756 @cindex @samp{qTfSTM} packet
40757 @cindex @samp{qTsSTM} packet
40758 These packets request data about static tracepoint markers that exist
40759 in the target program. @value{GDBN} sends @code{qTfSTM} to get the
40760 first piece of data, and multiple @code{qTsSTM} to get additional
40761 pieces. Replies to these packets take the following form:
40765 @item m @var{address}:@var{id}:@var{extra}
40767 @item m @var{address}:@var{id}:@var{extra},@var{address}:@var{id}:@var{extra}@dots{}
40768 a comma-separated list of markers
40770 (lower case letter @samp{L}) denotes end of list.
40772 An error occurred. The error number @var{nn} is given as hex digits.
40774 An empty reply indicates that the request is not supported by the
40778 The @var{address} is encoded in hex;
40779 @var{id} and @var{extra} are strings encoded in hex.
40781 In response to each query, the target will reply with a list of one or
40782 more markers, separated by commas. @value{GDBN} will respond to each
40783 reply with a request for more markers (using the @samp{qs} form of the
40784 query), until the target responds with @samp{l} (lower-case ell, for
40787 @item qTSTMat:@var{address}
40789 @cindex @samp{qTSTMat} packet
40790 This packets requests data about static tracepoint markers in the
40791 target program at @var{address}. Replies to this packet follow the
40792 syntax of the @samp{qTfSTM} and @code{qTsSTM} packets that list static
40793 tracepoint markers.
40795 @item QTSave:@var{filename}
40796 @cindex @samp{QTSave} packet
40797 This packet directs the target to save trace data to the file name
40798 @var{filename} in the target's filesystem. The @var{filename} is encoded
40799 as a hex string; the interpretation of the file name (relative vs
40800 absolute, wild cards, etc) is up to the target.
40802 @item qTBuffer:@var{offset},@var{len}
40803 @cindex @samp{qTBuffer} packet
40804 Return up to @var{len} bytes of the current contents of trace buffer,
40805 starting at @var{offset}. The trace buffer is treated as if it were
40806 a contiguous collection of traceframes, as per the trace file format.
40807 The reply consists as many hex-encoded bytes as the target can deliver
40808 in a packet; it is not an error to return fewer than were asked for.
40809 A reply consisting of just @code{l} indicates that no bytes are
40812 @item QTBuffer:circular:@var{value}
40813 This packet directs the target to use a circular trace buffer if
40814 @var{value} is 1, or a linear buffer if the value is 0.
40816 @item QTBuffer:size:@var{size}
40817 @anchor{QTBuffer-size}
40818 @cindex @samp{QTBuffer size} packet
40819 This packet directs the target to make the trace buffer be of size
40820 @var{size} if possible. A value of @code{-1} tells the target to
40821 use whatever size it prefers.
40823 @item QTNotes:@r{[}@var{type}:@var{text}@r{]}@r{[};@var{type}:@var{text}@r{]}@dots{}
40824 @cindex @samp{QTNotes} packet
40825 This packet adds optional textual notes to the trace run. Allowable
40826 types include @code{user}, @code{notes}, and @code{tstop}, the
40827 @var{text} fields are arbitrary strings, hex-encoded.
40831 @subsection Relocate instruction reply packet
40832 When installing fast tracepoints in memory, the target may need to
40833 relocate the instruction currently at the tracepoint address to a
40834 different address in memory. For most instructions, a simple copy is
40835 enough, but, for example, call instructions that implicitly push the
40836 return address on the stack, and relative branches or other
40837 PC-relative instructions require offset adjustment, so that the effect
40838 of executing the instruction at a different address is the same as if
40839 it had executed in the original location.
40841 In response to several of the tracepoint packets, the target may also
40842 respond with a number of intermediate @samp{qRelocInsn} request
40843 packets before the final result packet, to have @value{GDBN} handle
40844 this relocation operation. If a packet supports this mechanism, its
40845 documentation will explicitly say so. See for example the above
40846 descriptions for the @samp{QTStart} and @samp{QTDP} packets. The
40847 format of the request is:
40850 @item qRelocInsn:@var{from};@var{to}
40852 This requests @value{GDBN} to copy instruction at address @var{from}
40853 to address @var{to}, possibly adjusted so that executing the
40854 instruction at @var{to} has the same effect as executing it at
40855 @var{from}. @value{GDBN} writes the adjusted instruction to target
40856 memory starting at @var{to}.
40861 @item qRelocInsn:@var{adjusted_size}
40862 Informs the stub the relocation is complete. The @var{adjusted_size} is
40863 the length in bytes of resulting relocated instruction sequence.
40865 A badly formed request was detected, or an error was encountered while
40866 relocating the instruction.
40869 @node Host I/O Packets
40870 @section Host I/O Packets
40871 @cindex Host I/O, remote protocol
40872 @cindex file transfer, remote protocol
40874 The @dfn{Host I/O} packets allow @value{GDBN} to perform I/O
40875 operations on the far side of a remote link. For example, Host I/O is
40876 used to upload and download files to a remote target with its own
40877 filesystem. Host I/O uses the same constant values and data structure
40878 layout as the target-initiated File-I/O protocol. However, the
40879 Host I/O packets are structured differently. The target-initiated
40880 protocol relies on target memory to store parameters and buffers.
40881 Host I/O requests are initiated by @value{GDBN}, and the
40882 target's memory is not involved. @xref{File-I/O Remote Protocol
40883 Extension}, for more details on the target-initiated protocol.
40885 The Host I/O request packets all encode a single operation along with
40886 its arguments. They have this format:
40890 @item vFile:@var{operation}: @var{parameter}@dots{}
40891 @var{operation} is the name of the particular request; the target
40892 should compare the entire packet name up to the second colon when checking
40893 for a supported operation. The format of @var{parameter} depends on
40894 the operation. Numbers are always passed in hexadecimal. Negative
40895 numbers have an explicit minus sign (i.e.@: two's complement is not
40896 used). Strings (e.g.@: filenames) are encoded as a series of
40897 hexadecimal bytes. The last argument to a system call may be a
40898 buffer of escaped binary data (@pxref{Binary Data}).
40902 The valid responses to Host I/O packets are:
40906 @item F @var{result} [, @var{errno}] [; @var{attachment}]
40907 @var{result} is the integer value returned by this operation, usually
40908 non-negative for success and -1 for errors. If an error has occured,
40909 @var{errno} will be included in the result specifying a
40910 value defined by the File-I/O protocol (@pxref{Errno Values}). For
40911 operations which return data, @var{attachment} supplies the data as a
40912 binary buffer. Binary buffers in response packets are escaped in the
40913 normal way (@pxref{Binary Data}). See the individual packet
40914 documentation for the interpretation of @var{result} and
40918 An empty response indicates that this operation is not recognized.
40922 These are the supported Host I/O operations:
40925 @item vFile:open: @var{filename}, @var{flags}, @var{mode}
40926 Open a file at @var{filename} and return a file descriptor for it, or
40927 return -1 if an error occurs. The @var{filename} is a string,
40928 @var{flags} is an integer indicating a mask of open flags
40929 (@pxref{Open Flags}), and @var{mode} is an integer indicating a mask
40930 of mode bits to use if the file is created (@pxref{mode_t Values}).
40931 @xref{open}, for details of the open flags and mode values.
40933 @item vFile:close: @var{fd}
40934 Close the open file corresponding to @var{fd} and return 0, or
40935 -1 if an error occurs.
40937 @item vFile:pread: @var{fd}, @var{count}, @var{offset}
40938 Read data from the open file corresponding to @var{fd}. Up to
40939 @var{count} bytes will be read from the file, starting at @var{offset}
40940 relative to the start of the file. The target may read fewer bytes;
40941 common reasons include packet size limits and an end-of-file
40942 condition. The number of bytes read is returned. Zero should only be
40943 returned for a successful read at the end of the file, or if
40944 @var{count} was zero.
40946 The data read should be returned as a binary attachment on success.
40947 If zero bytes were read, the response should include an empty binary
40948 attachment (i.e.@: a trailing semicolon). The return value is the
40949 number of target bytes read; the binary attachment may be longer if
40950 some characters were escaped.
40952 @item vFile:pwrite: @var{fd}, @var{offset}, @var{data}
40953 Write @var{data} (a binary buffer) to the open file corresponding
40954 to @var{fd}. Start the write at @var{offset} from the start of the
40955 file. Unlike many @code{write} system calls, there is no
40956 separate @var{count} argument; the length of @var{data} in the
40957 packet is used. @samp{vFile:write} returns the number of bytes written,
40958 which may be shorter than the length of @var{data}, or -1 if an
40961 @item vFile:fstat: @var{fd}
40962 Get information about the open file corresponding to @var{fd}.
40963 On success the information is returned as a binary attachment
40964 and the return value is the size of this attachment in bytes.
40965 If an error occurs the return value is -1. The format of the
40966 returned binary attachment is as described in @ref{struct stat}.
40968 @item vFile:unlink: @var{filename}
40969 Delete the file at @var{filename} on the target. Return 0,
40970 or -1 if an error occurs. The @var{filename} is a string.
40972 @item vFile:readlink: @var{filename}
40973 Read value of symbolic link @var{filename} on the target. Return
40974 the number of bytes read, or -1 if an error occurs.
40976 The data read should be returned as a binary attachment on success.
40977 If zero bytes were read, the response should include an empty binary
40978 attachment (i.e.@: a trailing semicolon). The return value is the
40979 number of target bytes read; the binary attachment may be longer if
40980 some characters were escaped.
40982 @item vFile:setfs: @var{pid}
40983 Select the filesystem on which @code{vFile} operations with
40984 @var{filename} arguments will operate. This is required for
40985 @value{GDBN} to be able to access files on remote targets where
40986 the remote stub does not share a common filesystem with the
40989 If @var{pid} is nonzero, select the filesystem as seen by process
40990 @var{pid}. If @var{pid} is zero, select the filesystem as seen by
40991 the remote stub. Return 0 on success, or -1 if an error occurs.
40992 If @code{vFile:setfs:} indicates success, the selected filesystem
40993 remains selected until the next successful @code{vFile:setfs:}
40999 @section Interrupts
41000 @cindex interrupts (remote protocol)
41001 @anchor{interrupting remote targets}
41003 In all-stop mode, when a program on the remote target is running,
41004 @value{GDBN} may attempt to interrupt it by sending a @samp{Ctrl-C},
41005 @code{BREAK} or a @code{BREAK} followed by @code{g}, control of which
41006 is specified via @value{GDBN}'s @samp{interrupt-sequence}.
41008 The precise meaning of @code{BREAK} is defined by the transport
41009 mechanism and may, in fact, be undefined. @value{GDBN} does not
41010 currently define a @code{BREAK} mechanism for any of the network
41011 interfaces except for TCP, in which case @value{GDBN} sends the
41012 @code{telnet} BREAK sequence.
41014 @samp{Ctrl-C}, on the other hand, is defined and implemented for all
41015 transport mechanisms. It is represented by sending the single byte
41016 @code{0x03} without any of the usual packet overhead described in
41017 the Overview section (@pxref{Overview}). When a @code{0x03} byte is
41018 transmitted as part of a packet, it is considered to be packet data
41019 and does @emph{not} represent an interrupt. E.g., an @samp{X} packet
41020 (@pxref{X packet}), used for binary downloads, may include an unescaped
41021 @code{0x03} as part of its packet.
41023 @code{BREAK} followed by @code{g} is also known as Magic SysRq g.
41024 When Linux kernel receives this sequence from serial port,
41025 it stops execution and connects to gdb.
41027 In non-stop mode, because packet resumptions are asynchronous
41028 (@pxref{vCont packet}), @value{GDBN} is always free to send a remote
41029 command to the remote stub, even when the target is running. For that
41030 reason, @value{GDBN} instead sends a regular packet (@pxref{vCtrlC
41031 packet}) with the usual packet framing instead of the single byte
41034 Stubs are not required to recognize these interrupt mechanisms and the
41035 precise meaning associated with receipt of the interrupt is
41036 implementation defined. If the target supports debugging of multiple
41037 threads and/or processes, it should attempt to interrupt all
41038 currently-executing threads and processes.
41039 If the stub is successful at interrupting the
41040 running program, it should send one of the stop
41041 reply packets (@pxref{Stop Reply Packets}) to @value{GDBN} as a result
41042 of successfully stopping the program in all-stop mode, and a stop reply
41043 for each stopped thread in non-stop mode.
41044 Interrupts received while the
41045 program is stopped are queued and the program will be interrupted when
41046 it is resumed next time.
41048 @node Notification Packets
41049 @section Notification Packets
41050 @cindex notification packets
41051 @cindex packets, notification
41053 The @value{GDBN} remote serial protocol includes @dfn{notifications},
41054 packets that require no acknowledgment. Both the GDB and the stub
41055 may send notifications (although the only notifications defined at
41056 present are sent by the stub). Notifications carry information
41057 without incurring the round-trip latency of an acknowledgment, and so
41058 are useful for low-impact communications where occasional packet loss
41061 A notification packet has the form @samp{% @var{data} #
41062 @var{checksum}}, where @var{data} is the content of the notification,
41063 and @var{checksum} is a checksum of @var{data}, computed and formatted
41064 as for ordinary @value{GDBN} packets. A notification's @var{data}
41065 never contains @samp{$}, @samp{%} or @samp{#} characters. Upon
41066 receiving a notification, the recipient sends no @samp{+} or @samp{-}
41067 to acknowledge the notification's receipt or to report its corruption.
41069 Every notification's @var{data} begins with a name, which contains no
41070 colon characters, followed by a colon character.
41072 Recipients should silently ignore corrupted notifications and
41073 notifications they do not understand. Recipients should restart
41074 timeout periods on receipt of a well-formed notification, whether or
41075 not they understand it.
41077 Senders should only send the notifications described here when this
41078 protocol description specifies that they are permitted. In the
41079 future, we may extend the protocol to permit existing notifications in
41080 new contexts; this rule helps older senders avoid confusing newer
41083 (Older versions of @value{GDBN} ignore bytes received until they see
41084 the @samp{$} byte that begins an ordinary packet, so new stubs may
41085 transmit notifications without fear of confusing older clients. There
41086 are no notifications defined for @value{GDBN} to send at the moment, but we
41087 assume that most older stubs would ignore them, as well.)
41089 Each notification is comprised of three parts:
41091 @item @var{name}:@var{event}
41092 The notification packet is sent by the side that initiates the
41093 exchange (currently, only the stub does that), with @var{event}
41094 carrying the specific information about the notification, and
41095 @var{name} specifying the name of the notification.
41097 The acknowledge sent by the other side, usually @value{GDBN}, to
41098 acknowledge the exchange and request the event.
41101 The purpose of an asynchronous notification mechanism is to report to
41102 @value{GDBN} that something interesting happened in the remote stub.
41104 The remote stub may send notification @var{name}:@var{event}
41105 at any time, but @value{GDBN} acknowledges the notification when
41106 appropriate. The notification event is pending before @value{GDBN}
41107 acknowledges. Only one notification at a time may be pending; if
41108 additional events occur before @value{GDBN} has acknowledged the
41109 previous notification, they must be queued by the stub for later
41110 synchronous transmission in response to @var{ack} packets from
41111 @value{GDBN}. Because the notification mechanism is unreliable,
41112 the stub is permitted to resend a notification if it believes
41113 @value{GDBN} may not have received it.
41115 Specifically, notifications may appear when @value{GDBN} is not
41116 otherwise reading input from the stub, or when @value{GDBN} is
41117 expecting to read a normal synchronous response or a
41118 @samp{+}/@samp{-} acknowledgment to a packet it has sent.
41119 Notification packets are distinct from any other communication from
41120 the stub so there is no ambiguity.
41122 After receiving a notification, @value{GDBN} shall acknowledge it by
41123 sending a @var{ack} packet as a regular, synchronous request to the
41124 stub. Such acknowledgment is not required to happen immediately, as
41125 @value{GDBN} is permitted to send other, unrelated packets to the
41126 stub first, which the stub should process normally.
41128 Upon receiving a @var{ack} packet, if the stub has other queued
41129 events to report to @value{GDBN}, it shall respond by sending a
41130 normal @var{event}. @value{GDBN} shall then send another @var{ack}
41131 packet to solicit further responses; again, it is permitted to send
41132 other, unrelated packets as well which the stub should process
41135 If the stub receives a @var{ack} packet and there are no additional
41136 @var{event} to report, the stub shall return an @samp{OK} response.
41137 At this point, @value{GDBN} has finished processing a notification
41138 and the stub has completed sending any queued events. @value{GDBN}
41139 won't accept any new notifications until the final @samp{OK} is
41140 received . If further notification events occur, the stub shall send
41141 a new notification, @value{GDBN} shall accept the notification, and
41142 the process shall be repeated.
41144 The process of asynchronous notification can be illustrated by the
41147 <- @code{%Stop:T0505:98e7ffbf;04:4ce6ffbf;08:b1b6e54c;thread:p7526.7526;core:0;}
41150 <- @code{T0505:68f37db7;04:40f37db7;08:63850408;thread:p7526.7528;core:0;}
41152 <- @code{T0505:68e3fdb6;04:40e3fdb6;08:63850408;thread:p7526.7529;core:0;}
41157 The following notifications are defined:
41158 @multitable @columnfractions 0.12 0.12 0.38 0.38
41167 @tab @var{reply}. The @var{reply} has the form of a stop reply, as
41168 described in @ref{Stop Reply Packets}. Refer to @ref{Remote Non-Stop},
41169 for information on how these notifications are acknowledged by
41171 @tab Report an asynchronous stop event in non-stop mode.
41175 @node Remote Non-Stop
41176 @section Remote Protocol Support for Non-Stop Mode
41178 @value{GDBN}'s remote protocol supports non-stop debugging of
41179 multi-threaded programs, as described in @ref{Non-Stop Mode}. If the stub
41180 supports non-stop mode, it should report that to @value{GDBN} by including
41181 @samp{QNonStop+} in its @samp{qSupported} response (@pxref{qSupported}).
41183 @value{GDBN} typically sends a @samp{QNonStop} packet only when
41184 establishing a new connection with the stub. Entering non-stop mode
41185 does not alter the state of any currently-running threads, but targets
41186 must stop all threads in any already-attached processes when entering
41187 all-stop mode. @value{GDBN} uses the @samp{?} packet as necessary to
41188 probe the target state after a mode change.
41190 In non-stop mode, when an attached process encounters an event that
41191 would otherwise be reported with a stop reply, it uses the
41192 asynchronous notification mechanism (@pxref{Notification Packets}) to
41193 inform @value{GDBN}. In contrast to all-stop mode, where all threads
41194 in all processes are stopped when a stop reply is sent, in non-stop
41195 mode only the thread reporting the stop event is stopped. That is,
41196 when reporting a @samp{S} or @samp{T} response to indicate completion
41197 of a step operation, hitting a breakpoint, or a fault, only the
41198 affected thread is stopped; any other still-running threads continue
41199 to run. When reporting a @samp{W} or @samp{X} response, all running
41200 threads belonging to other attached processes continue to run.
41202 In non-stop mode, the target shall respond to the @samp{?} packet as
41203 follows. First, any incomplete stop reply notification/@samp{vStopped}
41204 sequence in progress is abandoned. The target must begin a new
41205 sequence reporting stop events for all stopped threads, whether or not
41206 it has previously reported those events to @value{GDBN}. The first
41207 stop reply is sent as a synchronous reply to the @samp{?} packet, and
41208 subsequent stop replies are sent as responses to @samp{vStopped} packets
41209 using the mechanism described above. The target must not send
41210 asynchronous stop reply notifications until the sequence is complete.
41211 If all threads are running when the target receives the @samp{?} packet,
41212 or if the target is not attached to any process, it shall respond
41215 If the stub supports non-stop mode, it should also support the
41216 @samp{swbreak} stop reason if software breakpoints are supported, and
41217 the @samp{hwbreak} stop reason if hardware breakpoints are supported
41218 (@pxref{swbreak stop reason}). This is because given the asynchronous
41219 nature of non-stop mode, between the time a thread hits a breakpoint
41220 and the time the event is finally processed by @value{GDBN}, the
41221 breakpoint may have already been removed from the target. Due to
41222 this, @value{GDBN} needs to be able to tell whether a trap stop was
41223 caused by a delayed breakpoint event, which should be ignored, as
41224 opposed to a random trap signal, which should be reported to the user.
41225 Note the @samp{swbreak} feature implies that the target is responsible
41226 for adjusting the PC when a software breakpoint triggers, if
41227 necessary, such as on the x86 architecture.
41229 @node Packet Acknowledgment
41230 @section Packet Acknowledgment
41232 @cindex acknowledgment, for @value{GDBN} remote
41233 @cindex packet acknowledgment, for @value{GDBN} remote
41234 By default, when either the host or the target machine receives a packet,
41235 the first response expected is an acknowledgment: either @samp{+} (to indicate
41236 the package was received correctly) or @samp{-} (to request retransmission).
41237 This mechanism allows the @value{GDBN} remote protocol to operate over
41238 unreliable transport mechanisms, such as a serial line.
41240 In cases where the transport mechanism is itself reliable (such as a pipe or
41241 TCP connection), the @samp{+}/@samp{-} acknowledgments are redundant.
41242 It may be desirable to disable them in that case to reduce communication
41243 overhead, or for other reasons. This can be accomplished by means of the
41244 @samp{QStartNoAckMode} packet; @pxref{QStartNoAckMode}.
41246 When in no-acknowledgment mode, neither the stub nor @value{GDBN} shall send or
41247 expect @samp{+}/@samp{-} protocol acknowledgments. The packet
41248 and response format still includes the normal checksum, as described in
41249 @ref{Overview}, but the checksum may be ignored by the receiver.
41251 If the stub supports @samp{QStartNoAckMode} and prefers to operate in
41252 no-acknowledgment mode, it should report that to @value{GDBN}
41253 by including @samp{QStartNoAckMode+} in its response to @samp{qSupported};
41254 @pxref{qSupported}.
41255 If @value{GDBN} also supports @samp{QStartNoAckMode} and it has not been
41256 disabled via the @code{set remote noack-packet off} command
41257 (@pxref{Remote Configuration}),
41258 @value{GDBN} may then send a @samp{QStartNoAckMode} packet to the stub.
41259 Only then may the stub actually turn off packet acknowledgments.
41260 @value{GDBN} sends a final @samp{+} acknowledgment of the stub's @samp{OK}
41261 response, which can be safely ignored by the stub.
41263 Note that @code{set remote noack-packet} command only affects negotiation
41264 between @value{GDBN} and the stub when subsequent connections are made;
41265 it does not affect the protocol acknowledgment state for any current
41267 Since @samp{+}/@samp{-} acknowledgments are enabled by default when a
41268 new connection is established,
41269 there is also no protocol request to re-enable the acknowledgments
41270 for the current connection, once disabled.
41275 Example sequence of a target being re-started. Notice how the restart
41276 does not get any direct output:
41281 @emph{target restarts}
41284 <- @code{T001:1234123412341234}
41288 Example sequence of a target being stepped by a single instruction:
41291 -> @code{G1445@dots{}}
41296 <- @code{T001:1234123412341234}
41300 <- @code{1455@dots{}}
41304 @node File-I/O Remote Protocol Extension
41305 @section File-I/O Remote Protocol Extension
41306 @cindex File-I/O remote protocol extension
41309 * File-I/O Overview::
41310 * Protocol Basics::
41311 * The F Request Packet::
41312 * The F Reply Packet::
41313 * The Ctrl-C Message::
41315 * List of Supported Calls::
41316 * Protocol-specific Representation of Datatypes::
41318 * File-I/O Examples::
41321 @node File-I/O Overview
41322 @subsection File-I/O Overview
41323 @cindex file-i/o overview
41325 The @dfn{File I/O remote protocol extension} (short: File-I/O) allows the
41326 target to use the host's file system and console I/O to perform various
41327 system calls. System calls on the target system are translated into a
41328 remote protocol packet to the host system, which then performs the needed
41329 actions and returns a response packet to the target system.
41330 This simulates file system operations even on targets that lack file systems.
41332 The protocol is defined to be independent of both the host and target systems.
41333 It uses its own internal representation of datatypes and values. Both
41334 @value{GDBN} and the target's @value{GDBN} stub are responsible for
41335 translating the system-dependent value representations into the internal
41336 protocol representations when data is transmitted.
41338 The communication is synchronous. A system call is possible only when
41339 @value{GDBN} is waiting for a response from the @samp{C}, @samp{c}, @samp{S}
41340 or @samp{s} packets. While @value{GDBN} handles the request for a system call,
41341 the target is stopped to allow deterministic access to the target's
41342 memory. Therefore File-I/O is not interruptible by target signals. On
41343 the other hand, it is possible to interrupt File-I/O by a user interrupt
41344 (@samp{Ctrl-C}) within @value{GDBN}.
41346 The target's request to perform a host system call does not finish
41347 the latest @samp{C}, @samp{c}, @samp{S} or @samp{s} action. That means,
41348 after finishing the system call, the target returns to continuing the
41349 previous activity (continue, step). No additional continue or step
41350 request from @value{GDBN} is required.
41353 (@value{GDBP}) continue
41354 <- target requests 'system call X'
41355 target is stopped, @value{GDBN} executes system call
41356 -> @value{GDBN} returns result
41357 ... target continues, @value{GDBN} returns to wait for the target
41358 <- target hits breakpoint and sends a Txx packet
41361 The protocol only supports I/O on the console and to regular files on
41362 the host file system. Character or block special devices, pipes,
41363 named pipes, sockets or any other communication method on the host
41364 system are not supported by this protocol.
41366 File I/O is not supported in non-stop mode.
41368 @node Protocol Basics
41369 @subsection Protocol Basics
41370 @cindex protocol basics, file-i/o
41372 The File-I/O protocol uses the @code{F} packet as the request as well
41373 as reply packet. Since a File-I/O system call can only occur when
41374 @value{GDBN} is waiting for a response from the continuing or stepping target,
41375 the File-I/O request is a reply that @value{GDBN} has to expect as a result
41376 of a previous @samp{C}, @samp{c}, @samp{S} or @samp{s} packet.
41377 This @code{F} packet contains all information needed to allow @value{GDBN}
41378 to call the appropriate host system call:
41382 A unique identifier for the requested system call.
41385 All parameters to the system call. Pointers are given as addresses
41386 in the target memory address space. Pointers to strings are given as
41387 pointer/length pair. Numerical values are given as they are.
41388 Numerical control flags are given in a protocol-specific representation.
41392 At this point, @value{GDBN} has to perform the following actions.
41396 If the parameters include pointer values to data needed as input to a
41397 system call, @value{GDBN} requests this data from the target with a
41398 standard @code{m} packet request. This additional communication has to be
41399 expected by the target implementation and is handled as any other @code{m}
41403 @value{GDBN} translates all value from protocol representation to host
41404 representation as needed. Datatypes are coerced into the host types.
41407 @value{GDBN} calls the system call.
41410 It then coerces datatypes back to protocol representation.
41413 If the system call is expected to return data in buffer space specified
41414 by pointer parameters to the call, the data is transmitted to the
41415 target using a @code{M} or @code{X} packet. This packet has to be expected
41416 by the target implementation and is handled as any other @code{M} or @code{X}
41421 Eventually @value{GDBN} replies with another @code{F} packet which contains all
41422 necessary information for the target to continue. This at least contains
41429 @code{errno}, if has been changed by the system call.
41436 After having done the needed type and value coercion, the target continues
41437 the latest continue or step action.
41439 @node The F Request Packet
41440 @subsection The @code{F} Request Packet
41441 @cindex file-i/o request packet
41442 @cindex @code{F} request packet
41444 The @code{F} request packet has the following format:
41447 @item F@var{call-id},@var{parameter@dots{}}
41449 @var{call-id} is the identifier to indicate the host system call to be called.
41450 This is just the name of the function.
41452 @var{parameter@dots{}} are the parameters to the system call.
41453 Parameters are hexadecimal integer values, either the actual values in case
41454 of scalar datatypes, pointers to target buffer space in case of compound
41455 datatypes and unspecified memory areas, or pointer/length pairs in case
41456 of string parameters. These are appended to the @var{call-id} as a
41457 comma-delimited list. All values are transmitted in ASCII
41458 string representation, pointer/length pairs separated by a slash.
41464 @node The F Reply Packet
41465 @subsection The @code{F} Reply Packet
41466 @cindex file-i/o reply packet
41467 @cindex @code{F} reply packet
41469 The @code{F} reply packet has the following format:
41473 @item F@var{retcode},@var{errno},@var{Ctrl-C flag};@var{call-specific attachment}
41475 @var{retcode} is the return code of the system call as hexadecimal value.
41477 @var{errno} is the @code{errno} set by the call, in protocol-specific
41479 This parameter can be omitted if the call was successful.
41481 @var{Ctrl-C flag} is only sent if the user requested a break. In this
41482 case, @var{errno} must be sent as well, even if the call was successful.
41483 The @var{Ctrl-C flag} itself consists of the character @samp{C}:
41490 or, if the call was interrupted before the host call has been performed:
41497 assuming 4 is the protocol-specific representation of @code{EINTR}.
41502 @node The Ctrl-C Message
41503 @subsection The @samp{Ctrl-C} Message
41504 @cindex ctrl-c message, in file-i/o protocol
41506 If the @samp{Ctrl-C} flag is set in the @value{GDBN}
41507 reply packet (@pxref{The F Reply Packet}),
41508 the target should behave as if it had
41509 gotten a break message. The meaning for the target is ``system call
41510 interrupted by @code{SIGINT}''. Consequentially, the target should actually stop
41511 (as with a break message) and return to @value{GDBN} with a @code{T02}
41514 It's important for the target to know in which
41515 state the system call was interrupted. There are two possible cases:
41519 The system call hasn't been performed on the host yet.
41522 The system call on the host has been finished.
41526 These two states can be distinguished by the target by the value of the
41527 returned @code{errno}. If it's the protocol representation of @code{EINTR}, the system
41528 call hasn't been performed. This is equivalent to the @code{EINTR} handling
41529 on POSIX systems. In any other case, the target may presume that the
41530 system call has been finished --- successfully or not --- and should behave
41531 as if the break message arrived right after the system call.
41533 @value{GDBN} must behave reliably. If the system call has not been called
41534 yet, @value{GDBN} may send the @code{F} reply immediately, setting @code{EINTR} as
41535 @code{errno} in the packet. If the system call on the host has been finished
41536 before the user requests a break, the full action must be finished by
41537 @value{GDBN}. This requires sending @code{M} or @code{X} packets as necessary.
41538 The @code{F} packet may only be sent when either nothing has happened
41539 or the full action has been completed.
41542 @subsection Console I/O
41543 @cindex console i/o as part of file-i/o
41545 By default and if not explicitly closed by the target system, the file
41546 descriptors 0, 1 and 2 are connected to the @value{GDBN} console. Output
41547 on the @value{GDBN} console is handled as any other file output operation
41548 (@code{write(1, @dots{})} or @code{write(2, @dots{})}). Console input is handled
41549 by @value{GDBN} so that after the target read request from file descriptor
41550 0 all following typing is buffered until either one of the following
41555 The user types @kbd{Ctrl-c}. The behaviour is as explained above, and the
41557 system call is treated as finished.
41560 The user presses @key{RET}. This is treated as end of input with a trailing
41564 The user types @kbd{Ctrl-d}. This is treated as end of input. No trailing
41565 character (neither newline nor @samp{Ctrl-D}) is appended to the input.
41569 If the user has typed more characters than fit in the buffer given to
41570 the @code{read} call, the trailing characters are buffered in @value{GDBN} until
41571 either another @code{read(0, @dots{})} is requested by the target, or debugging
41572 is stopped at the user's request.
41575 @node List of Supported Calls
41576 @subsection List of Supported Calls
41577 @cindex list of supported file-i/o calls
41594 @unnumberedsubsubsec open
41595 @cindex open, file-i/o system call
41600 int open(const char *pathname, int flags);
41601 int open(const char *pathname, int flags, mode_t mode);
41605 @samp{Fopen,@var{pathptr}/@var{len},@var{flags},@var{mode}}
41608 @var{flags} is the bitwise @code{OR} of the following values:
41612 If the file does not exist it will be created. The host
41613 rules apply as far as file ownership and time stamps
41617 When used with @code{O_CREAT}, if the file already exists it is
41618 an error and open() fails.
41621 If the file already exists and the open mode allows
41622 writing (@code{O_RDWR} or @code{O_WRONLY} is given) it will be
41623 truncated to zero length.
41626 The file is opened in append mode.
41629 The file is opened for reading only.
41632 The file is opened for writing only.
41635 The file is opened for reading and writing.
41639 Other bits are silently ignored.
41643 @var{mode} is the bitwise @code{OR} of the following values:
41647 User has read permission.
41650 User has write permission.
41653 Group has read permission.
41656 Group has write permission.
41659 Others have read permission.
41662 Others have write permission.
41666 Other bits are silently ignored.
41669 @item Return value:
41670 @code{open} returns the new file descriptor or -1 if an error
41677 @var{pathname} already exists and @code{O_CREAT} and @code{O_EXCL} were used.
41680 @var{pathname} refers to a directory.
41683 The requested access is not allowed.
41686 @var{pathname} was too long.
41689 A directory component in @var{pathname} does not exist.
41692 @var{pathname} refers to a device, pipe, named pipe or socket.
41695 @var{pathname} refers to a file on a read-only filesystem and
41696 write access was requested.
41699 @var{pathname} is an invalid pointer value.
41702 No space on device to create the file.
41705 The process already has the maximum number of files open.
41708 The limit on the total number of files open on the system
41712 The call was interrupted by the user.
41718 @unnumberedsubsubsec close
41719 @cindex close, file-i/o system call
41728 @samp{Fclose,@var{fd}}
41730 @item Return value:
41731 @code{close} returns zero on success, or -1 if an error occurred.
41737 @var{fd} isn't a valid open file descriptor.
41740 The call was interrupted by the user.
41746 @unnumberedsubsubsec read
41747 @cindex read, file-i/o system call
41752 int read(int fd, void *buf, unsigned int count);
41756 @samp{Fread,@var{fd},@var{bufptr},@var{count}}
41758 @item Return value:
41759 On success, the number of bytes read is returned.
41760 Zero indicates end of file. If count is zero, read
41761 returns zero as well. On error, -1 is returned.
41767 @var{fd} is not a valid file descriptor or is not open for
41771 @var{bufptr} is an invalid pointer value.
41774 The call was interrupted by the user.
41780 @unnumberedsubsubsec write
41781 @cindex write, file-i/o system call
41786 int write(int fd, const void *buf, unsigned int count);
41790 @samp{Fwrite,@var{fd},@var{bufptr},@var{count}}
41792 @item Return value:
41793 On success, the number of bytes written are returned.
41794 Zero indicates nothing was written. On error, -1
41801 @var{fd} is not a valid file descriptor or is not open for
41805 @var{bufptr} is an invalid pointer value.
41808 An attempt was made to write a file that exceeds the
41809 host-specific maximum file size allowed.
41812 No space on device to write the data.
41815 The call was interrupted by the user.
41821 @unnumberedsubsubsec lseek
41822 @cindex lseek, file-i/o system call
41827 long lseek (int fd, long offset, int flag);
41831 @samp{Flseek,@var{fd},@var{offset},@var{flag}}
41833 @var{flag} is one of:
41837 The offset is set to @var{offset} bytes.
41840 The offset is set to its current location plus @var{offset}
41844 The offset is set to the size of the file plus @var{offset}
41848 @item Return value:
41849 On success, the resulting unsigned offset in bytes from
41850 the beginning of the file is returned. Otherwise, a
41851 value of -1 is returned.
41857 @var{fd} is not a valid open file descriptor.
41860 @var{fd} is associated with the @value{GDBN} console.
41863 @var{flag} is not a proper value.
41866 The call was interrupted by the user.
41872 @unnumberedsubsubsec rename
41873 @cindex rename, file-i/o system call
41878 int rename(const char *oldpath, const char *newpath);
41882 @samp{Frename,@var{oldpathptr}/@var{len},@var{newpathptr}/@var{len}}
41884 @item Return value:
41885 On success, zero is returned. On error, -1 is returned.
41891 @var{newpath} is an existing directory, but @var{oldpath} is not a
41895 @var{newpath} is a non-empty directory.
41898 @var{oldpath} or @var{newpath} is a directory that is in use by some
41902 An attempt was made to make a directory a subdirectory
41906 A component used as a directory in @var{oldpath} or new
41907 path is not a directory. Or @var{oldpath} is a directory
41908 and @var{newpath} exists but is not a directory.
41911 @var{oldpathptr} or @var{newpathptr} are invalid pointer values.
41914 No access to the file or the path of the file.
41918 @var{oldpath} or @var{newpath} was too long.
41921 A directory component in @var{oldpath} or @var{newpath} does not exist.
41924 The file is on a read-only filesystem.
41927 The device containing the file has no room for the new
41931 The call was interrupted by the user.
41937 @unnumberedsubsubsec unlink
41938 @cindex unlink, file-i/o system call
41943 int unlink(const char *pathname);
41947 @samp{Funlink,@var{pathnameptr}/@var{len}}
41949 @item Return value:
41950 On success, zero is returned. On error, -1 is returned.
41956 No access to the file or the path of the file.
41959 The system does not allow unlinking of directories.
41962 The file @var{pathname} cannot be unlinked because it's
41963 being used by another process.
41966 @var{pathnameptr} is an invalid pointer value.
41969 @var{pathname} was too long.
41972 A directory component in @var{pathname} does not exist.
41975 A component of the path is not a directory.
41978 The file is on a read-only filesystem.
41981 The call was interrupted by the user.
41987 @unnumberedsubsubsec stat/fstat
41988 @cindex fstat, file-i/o system call
41989 @cindex stat, file-i/o system call
41994 int stat(const char *pathname, struct stat *buf);
41995 int fstat(int fd, struct stat *buf);
41999 @samp{Fstat,@var{pathnameptr}/@var{len},@var{bufptr}}@*
42000 @samp{Ffstat,@var{fd},@var{bufptr}}
42002 @item Return value:
42003 On success, zero is returned. On error, -1 is returned.
42009 @var{fd} is not a valid open file.
42012 A directory component in @var{pathname} does not exist or the
42013 path is an empty string.
42016 A component of the path is not a directory.
42019 @var{pathnameptr} is an invalid pointer value.
42022 No access to the file or the path of the file.
42025 @var{pathname} was too long.
42028 The call was interrupted by the user.
42034 @unnumberedsubsubsec gettimeofday
42035 @cindex gettimeofday, file-i/o system call
42040 int gettimeofday(struct timeval *tv, void *tz);
42044 @samp{Fgettimeofday,@var{tvptr},@var{tzptr}}
42046 @item Return value:
42047 On success, 0 is returned, -1 otherwise.
42053 @var{tz} is a non-NULL pointer.
42056 @var{tvptr} and/or @var{tzptr} is an invalid pointer value.
42062 @unnumberedsubsubsec isatty
42063 @cindex isatty, file-i/o system call
42068 int isatty(int fd);
42072 @samp{Fisatty,@var{fd}}
42074 @item Return value:
42075 Returns 1 if @var{fd} refers to the @value{GDBN} console, 0 otherwise.
42081 The call was interrupted by the user.
42086 Note that the @code{isatty} call is treated as a special case: it returns
42087 1 to the target if the file descriptor is attached
42088 to the @value{GDBN} console, 0 otherwise. Implementing through system calls
42089 would require implementing @code{ioctl} and would be more complex than
42094 @unnumberedsubsubsec system
42095 @cindex system, file-i/o system call
42100 int system(const char *command);
42104 @samp{Fsystem,@var{commandptr}/@var{len}}
42106 @item Return value:
42107 If @var{len} is zero, the return value indicates whether a shell is
42108 available. A zero return value indicates a shell is not available.
42109 For non-zero @var{len}, the value returned is -1 on error and the
42110 return status of the command otherwise. Only the exit status of the
42111 command is returned, which is extracted from the host's @code{system}
42112 return value by calling @code{WEXITSTATUS(retval)}. In case
42113 @file{/bin/sh} could not be executed, 127 is returned.
42119 The call was interrupted by the user.
42124 @value{GDBN} takes over the full task of calling the necessary host calls
42125 to perform the @code{system} call. The return value of @code{system} on
42126 the host is simplified before it's returned
42127 to the target. Any termination signal information from the child process
42128 is discarded, and the return value consists
42129 entirely of the exit status of the called command.
42131 Due to security concerns, the @code{system} call is by default refused
42132 by @value{GDBN}. The user has to allow this call explicitly with the
42133 @code{set remote system-call-allowed 1} command.
42136 @item set remote system-call-allowed
42137 @kindex set remote system-call-allowed
42138 Control whether to allow the @code{system} calls in the File I/O
42139 protocol for the remote target. The default is zero (disabled).
42141 @item show remote system-call-allowed
42142 @kindex show remote system-call-allowed
42143 Show whether the @code{system} calls are allowed in the File I/O
42147 @node Protocol-specific Representation of Datatypes
42148 @subsection Protocol-specific Representation of Datatypes
42149 @cindex protocol-specific representation of datatypes, in file-i/o protocol
42152 * Integral Datatypes::
42154 * Memory Transfer::
42159 @node Integral Datatypes
42160 @unnumberedsubsubsec Integral Datatypes
42161 @cindex integral datatypes, in file-i/o protocol
42163 The integral datatypes used in the system calls are @code{int},
42164 @code{unsigned int}, @code{long}, @code{unsigned long},
42165 @code{mode_t}, and @code{time_t}.
42167 @code{int}, @code{unsigned int}, @code{mode_t} and @code{time_t} are
42168 implemented as 32 bit values in this protocol.
42170 @code{long} and @code{unsigned long} are implemented as 64 bit types.
42172 @xref{Limits}, for corresponding MIN and MAX values (similar to those
42173 in @file{limits.h}) to allow range checking on host and target.
42175 @code{time_t} datatypes are defined as seconds since the Epoch.
42177 All integral datatypes transferred as part of a memory read or write of a
42178 structured datatype e.g.@: a @code{struct stat} have to be given in big endian
42181 @node Pointer Values
42182 @unnumberedsubsubsec Pointer Values
42183 @cindex pointer values, in file-i/o protocol
42185 Pointers to target data are transmitted as they are. An exception
42186 is made for pointers to buffers for which the length isn't
42187 transmitted as part of the function call, namely strings. Strings
42188 are transmitted as a pointer/length pair, both as hex values, e.g.@:
42195 which is a pointer to data of length 18 bytes at position 0x1aaf.
42196 The length is defined as the full string length in bytes, including
42197 the trailing null byte. For example, the string @code{"hello world"}
42198 at address 0x123456 is transmitted as
42204 @node Memory Transfer
42205 @unnumberedsubsubsec Memory Transfer
42206 @cindex memory transfer, in file-i/o protocol
42208 Structured data which is transferred using a memory read or write (for
42209 example, a @code{struct stat}) is expected to be in a protocol-specific format
42210 with all scalar multibyte datatypes being big endian. Translation to
42211 this representation needs to be done both by the target before the @code{F}
42212 packet is sent, and by @value{GDBN} before
42213 it transfers memory to the target. Transferred pointers to structured
42214 data should point to the already-coerced data at any time.
42218 @unnumberedsubsubsec struct stat
42219 @cindex struct stat, in file-i/o protocol
42221 The buffer of type @code{struct stat} used by the target and @value{GDBN}
42222 is defined as follows:
42226 unsigned int st_dev; /* device */
42227 unsigned int st_ino; /* inode */
42228 mode_t st_mode; /* protection */
42229 unsigned int st_nlink; /* number of hard links */
42230 unsigned int st_uid; /* user ID of owner */
42231 unsigned int st_gid; /* group ID of owner */
42232 unsigned int st_rdev; /* device type (if inode device) */
42233 unsigned long st_size; /* total size, in bytes */
42234 unsigned long st_blksize; /* blocksize for filesystem I/O */
42235 unsigned long st_blocks; /* number of blocks allocated */
42236 time_t st_atime; /* time of last access */
42237 time_t st_mtime; /* time of last modification */
42238 time_t st_ctime; /* time of last change */
42242 The integral datatypes conform to the definitions given in the
42243 appropriate section (see @ref{Integral Datatypes}, for details) so this
42244 structure is of size 64 bytes.
42246 The values of several fields have a restricted meaning and/or
42252 A value of 0 represents a file, 1 the console.
42255 No valid meaning for the target. Transmitted unchanged.
42258 Valid mode bits are described in @ref{Constants}. Any other
42259 bits have currently no meaning for the target.
42264 No valid meaning for the target. Transmitted unchanged.
42269 These values have a host and file system dependent
42270 accuracy. Especially on Windows hosts, the file system may not
42271 support exact timing values.
42274 The target gets a @code{struct stat} of the above representation and is
42275 responsible for coercing it to the target representation before
42278 Note that due to size differences between the host, target, and protocol
42279 representations of @code{struct stat} members, these members could eventually
42280 get truncated on the target.
42282 @node struct timeval
42283 @unnumberedsubsubsec struct timeval
42284 @cindex struct timeval, in file-i/o protocol
42286 The buffer of type @code{struct timeval} used by the File-I/O protocol
42287 is defined as follows:
42291 time_t tv_sec; /* second */
42292 long tv_usec; /* microsecond */
42296 The integral datatypes conform to the definitions given in the
42297 appropriate section (see @ref{Integral Datatypes}, for details) so this
42298 structure is of size 8 bytes.
42301 @subsection Constants
42302 @cindex constants, in file-i/o protocol
42304 The following values are used for the constants inside of the
42305 protocol. @value{GDBN} and target are responsible for translating these
42306 values before and after the call as needed.
42317 @unnumberedsubsubsec Open Flags
42318 @cindex open flags, in file-i/o protocol
42320 All values are given in hexadecimal representation.
42332 @node mode_t Values
42333 @unnumberedsubsubsec mode_t Values
42334 @cindex mode_t values, in file-i/o protocol
42336 All values are given in octal representation.
42353 @unnumberedsubsubsec Errno Values
42354 @cindex errno values, in file-i/o protocol
42356 All values are given in decimal representation.
42381 @code{EUNKNOWN} is used as a fallback error value if a host system returns
42382 any error value not in the list of supported error numbers.
42385 @unnumberedsubsubsec Lseek Flags
42386 @cindex lseek flags, in file-i/o protocol
42395 @unnumberedsubsubsec Limits
42396 @cindex limits, in file-i/o protocol
42398 All values are given in decimal representation.
42401 INT_MIN -2147483648
42403 UINT_MAX 4294967295
42404 LONG_MIN -9223372036854775808
42405 LONG_MAX 9223372036854775807
42406 ULONG_MAX 18446744073709551615
42409 @node File-I/O Examples
42410 @subsection File-I/O Examples
42411 @cindex file-i/o examples
42413 Example sequence of a write call, file descriptor 3, buffer is at target
42414 address 0x1234, 6 bytes should be written:
42417 <- @code{Fwrite,3,1234,6}
42418 @emph{request memory read from target}
42421 @emph{return "6 bytes written"}
42425 Example sequence of a read call, file descriptor 3, buffer is at target
42426 address 0x1234, 6 bytes should be read:
42429 <- @code{Fread,3,1234,6}
42430 @emph{request memory write to target}
42431 -> @code{X1234,6:XXXXXX}
42432 @emph{return "6 bytes read"}
42436 Example sequence of a read call, call fails on the host due to invalid
42437 file descriptor (@code{EBADF}):
42440 <- @code{Fread,3,1234,6}
42444 Example sequence of a read call, user presses @kbd{Ctrl-c} before syscall on
42448 <- @code{Fread,3,1234,6}
42453 Example sequence of a read call, user presses @kbd{Ctrl-c} after syscall on
42457 <- @code{Fread,3,1234,6}
42458 -> @code{X1234,6:XXXXXX}
42462 @node Library List Format
42463 @section Library List Format
42464 @cindex library list format, remote protocol
42466 On some platforms, a dynamic loader (e.g.@: @file{ld.so}) runs in the
42467 same process as your application to manage libraries. In this case,
42468 @value{GDBN} can use the loader's symbol table and normal memory
42469 operations to maintain a list of shared libraries. On other
42470 platforms, the operating system manages loaded libraries.
42471 @value{GDBN} can not retrieve the list of currently loaded libraries
42472 through memory operations, so it uses the @samp{qXfer:libraries:read}
42473 packet (@pxref{qXfer library list read}) instead. The remote stub
42474 queries the target's operating system and reports which libraries
42477 The @samp{qXfer:libraries:read} packet returns an XML document which
42478 lists loaded libraries and their offsets. Each library has an
42479 associated name and one or more segment or section base addresses,
42480 which report where the library was loaded in memory.
42482 For the common case of libraries that are fully linked binaries, the
42483 library should have a list of segments. If the target supports
42484 dynamic linking of a relocatable object file, its library XML element
42485 should instead include a list of allocated sections. The segment or
42486 section bases are start addresses, not relocation offsets; they do not
42487 depend on the library's link-time base addresses.
42489 @value{GDBN} must be linked with the Expat library to support XML
42490 library lists. @xref{Expat}.
42492 A simple memory map, with one loaded library relocated by a single
42493 offset, looks like this:
42497 <library name="/lib/libc.so.6">
42498 <segment address="0x10000000"/>
42503 Another simple memory map, with one loaded library with three
42504 allocated sections (.text, .data, .bss), looks like this:
42508 <library name="sharedlib.o">
42509 <section address="0x10000000"/>
42510 <section address="0x20000000"/>
42511 <section address="0x30000000"/>
42516 The format of a library list is described by this DTD:
42519 <!-- library-list: Root element with versioning -->
42520 <!ELEMENT library-list (library)*>
42521 <!ATTLIST library-list version CDATA #FIXED "1.0">
42522 <!ELEMENT library (segment*, section*)>
42523 <!ATTLIST library name CDATA #REQUIRED>
42524 <!ELEMENT segment EMPTY>
42525 <!ATTLIST segment address CDATA #REQUIRED>
42526 <!ELEMENT section EMPTY>
42527 <!ATTLIST section address CDATA #REQUIRED>
42530 In addition, segments and section descriptors cannot be mixed within a
42531 single library element, and you must supply at least one segment or
42532 section for each library.
42534 @node Library List Format for SVR4 Targets
42535 @section Library List Format for SVR4 Targets
42536 @cindex library list format, remote protocol
42538 On SVR4 platforms @value{GDBN} can use the symbol table of a dynamic loader
42539 (e.g.@: @file{ld.so}) and normal memory operations to maintain a list of
42540 shared libraries. Still a special library list provided by this packet is
42541 more efficient for the @value{GDBN} remote protocol.
42543 The @samp{qXfer:libraries-svr4:read} packet returns an XML document which lists
42544 loaded libraries and their SVR4 linker parameters. For each library on SVR4
42545 target, the following parameters are reported:
42549 @code{name}, the absolute file name from the @code{l_name} field of
42550 @code{struct link_map}.
42552 @code{lm} with address of @code{struct link_map} used for TLS
42553 (Thread Local Storage) access.
42555 @code{l_addr}, the displacement as read from the field @code{l_addr} of
42556 @code{struct link_map}. For prelinked libraries this is not an absolute
42557 memory address. It is a displacement of absolute memory address against
42558 address the file was prelinked to during the library load.
42560 @code{l_ld}, which is memory address of the @code{PT_DYNAMIC} segment
42563 Additionally the single @code{main-lm} attribute specifies address of
42564 @code{struct link_map} used for the main executable. This parameter is used
42565 for TLS access and its presence is optional.
42567 @value{GDBN} must be linked with the Expat library to support XML
42568 SVR4 library lists. @xref{Expat}.
42570 A simple memory map, with two loaded libraries (which do not use prelink),
42574 <library-list-svr4 version="1.0" main-lm="0xe4f8f8">
42575 <library name="/lib/ld-linux.so.2" lm="0xe4f51c" l_addr="0xe2d000"
42577 <library name="/lib/libc.so.6" lm="0xe4fbe8" l_addr="0x154000"
42579 </library-list-svr>
42582 The format of an SVR4 library list is described by this DTD:
42585 <!-- library-list-svr4: Root element with versioning -->
42586 <!ELEMENT library-list-svr4 (library)*>
42587 <!ATTLIST library-list-svr4 version CDATA #FIXED "1.0">
42588 <!ATTLIST library-list-svr4 main-lm CDATA #IMPLIED>
42589 <!ELEMENT library EMPTY>
42590 <!ATTLIST library name CDATA #REQUIRED>
42591 <!ATTLIST library lm CDATA #REQUIRED>
42592 <!ATTLIST library l_addr CDATA #REQUIRED>
42593 <!ATTLIST library l_ld CDATA #REQUIRED>
42596 @node Memory Map Format
42597 @section Memory Map Format
42598 @cindex memory map format
42600 To be able to write into flash memory, @value{GDBN} needs to obtain a
42601 memory map from the target. This section describes the format of the
42604 The memory map is obtained using the @samp{qXfer:memory-map:read}
42605 (@pxref{qXfer memory map read}) packet and is an XML document that
42606 lists memory regions.
42608 @value{GDBN} must be linked with the Expat library to support XML
42609 memory maps. @xref{Expat}.
42611 The top-level structure of the document is shown below:
42614 <?xml version="1.0"?>
42615 <!DOCTYPE memory-map
42616 PUBLIC "+//IDN gnu.org//DTD GDB Memory Map V1.0//EN"
42617 "http://sourceware.org/gdb/gdb-memory-map.dtd">
42623 Each region can be either:
42628 A region of RAM starting at @var{addr} and extending for @var{length}
42632 <memory type="ram" start="@var{addr}" length="@var{length}"/>
42637 A region of read-only memory:
42640 <memory type="rom" start="@var{addr}" length="@var{length}"/>
42645 A region of flash memory, with erasure blocks @var{blocksize}
42649 <memory type="flash" start="@var{addr}" length="@var{length}">
42650 <property name="blocksize">@var{blocksize}</property>
42656 Regions must not overlap. @value{GDBN} assumes that areas of memory not covered
42657 by the memory map are RAM, and uses the ordinary @samp{M} and @samp{X}
42658 packets to write to addresses in such ranges.
42660 The formal DTD for memory map format is given below:
42663 <!-- ................................................... -->
42664 <!-- Memory Map XML DTD ................................ -->
42665 <!-- File: memory-map.dtd .............................. -->
42666 <!-- .................................... .............. -->
42667 <!-- memory-map.dtd -->
42668 <!-- memory-map: Root element with versioning -->
42669 <!ELEMENT memory-map (memory)*>
42670 <!ATTLIST memory-map version CDATA #FIXED "1.0.0">
42671 <!ELEMENT memory (property)*>
42672 <!-- memory: Specifies a memory region,
42673 and its type, or device. -->
42674 <!ATTLIST memory type (ram|rom|flash) #REQUIRED
42675 start CDATA #REQUIRED
42676 length CDATA #REQUIRED>
42677 <!-- property: Generic attribute tag -->
42678 <!ELEMENT property (#PCDATA | property)*>
42679 <!ATTLIST property name (blocksize) #REQUIRED>
42682 @node Thread List Format
42683 @section Thread List Format
42684 @cindex thread list format
42686 To efficiently update the list of threads and their attributes,
42687 @value{GDBN} issues the @samp{qXfer:threads:read} packet
42688 (@pxref{qXfer threads read}) and obtains the XML document with
42689 the following structure:
42692 <?xml version="1.0"?>
42694 <thread id="id" core="0" name="name">
42695 ... description ...
42700 Each @samp{thread} element must have the @samp{id} attribute that
42701 identifies the thread (@pxref{thread-id syntax}). The
42702 @samp{core} attribute, if present, specifies which processor core
42703 the thread was last executing on. The @samp{name} attribute, if
42704 present, specifies the human-readable name of the thread. The content
42705 of the of @samp{thread} element is interpreted as human-readable
42706 auxiliary information. The @samp{handle} attribute, if present,
42707 is a hex encoded representation of the thread handle.
42710 @node Traceframe Info Format
42711 @section Traceframe Info Format
42712 @cindex traceframe info format
42714 To be able to know which objects in the inferior can be examined when
42715 inspecting a tracepoint hit, @value{GDBN} needs to obtain the list of
42716 memory ranges, registers and trace state variables that have been
42717 collected in a traceframe.
42719 This list is obtained using the @samp{qXfer:traceframe-info:read}
42720 (@pxref{qXfer traceframe info read}) packet and is an XML document.
42722 @value{GDBN} must be linked with the Expat library to support XML
42723 traceframe info discovery. @xref{Expat}.
42725 The top-level structure of the document is shown below:
42728 <?xml version="1.0"?>
42729 <!DOCTYPE traceframe-info
42730 PUBLIC "+//IDN gnu.org//DTD GDB Memory Map V1.0//EN"
42731 "http://sourceware.org/gdb/gdb-traceframe-info.dtd">
42737 Each traceframe block can be either:
42742 A region of collected memory starting at @var{addr} and extending for
42743 @var{length} bytes from there:
42746 <memory start="@var{addr}" length="@var{length}"/>
42750 A block indicating trace state variable numbered @var{number} has been
42754 <tvar id="@var{number}"/>
42759 The formal DTD for the traceframe info format is given below:
42762 <!ELEMENT traceframe-info (memory | tvar)* >
42763 <!ATTLIST traceframe-info version CDATA #FIXED "1.0">
42765 <!ELEMENT memory EMPTY>
42766 <!ATTLIST memory start CDATA #REQUIRED
42767 length CDATA #REQUIRED>
42769 <!ATTLIST tvar id CDATA #REQUIRED>
42772 @node Branch Trace Format
42773 @section Branch Trace Format
42774 @cindex branch trace format
42776 In order to display the branch trace of an inferior thread,
42777 @value{GDBN} needs to obtain the list of branches. This list is
42778 represented as list of sequential code blocks that are connected via
42779 branches. The code in each block has been executed sequentially.
42781 This list is obtained using the @samp{qXfer:btrace:read}
42782 (@pxref{qXfer btrace read}) packet and is an XML document.
42784 @value{GDBN} must be linked with the Expat library to support XML
42785 traceframe info discovery. @xref{Expat}.
42787 The top-level structure of the document is shown below:
42790 <?xml version="1.0"?>
42792 PUBLIC "+//IDN gnu.org//DTD GDB Branch Trace V1.0//EN"
42793 "http://sourceware.org/gdb/gdb-btrace.dtd">
42802 A block of sequentially executed instructions starting at @var{begin}
42803 and ending at @var{end}:
42806 <block begin="@var{begin}" end="@var{end}"/>
42811 The formal DTD for the branch trace format is given below:
42814 <!ELEMENT btrace (block* | pt) >
42815 <!ATTLIST btrace version CDATA #FIXED "1.0">
42817 <!ELEMENT block EMPTY>
42818 <!ATTLIST block begin CDATA #REQUIRED
42819 end CDATA #REQUIRED>
42821 <!ELEMENT pt (pt-config?, raw?)>
42823 <!ELEMENT pt-config (cpu?)>
42825 <!ELEMENT cpu EMPTY>
42826 <!ATTLIST cpu vendor CDATA #REQUIRED
42827 family CDATA #REQUIRED
42828 model CDATA #REQUIRED
42829 stepping CDATA #REQUIRED>
42831 <!ELEMENT raw (#PCDATA)>
42834 @node Branch Trace Configuration Format
42835 @section Branch Trace Configuration Format
42836 @cindex branch trace configuration format
42838 For each inferior thread, @value{GDBN} can obtain the branch trace
42839 configuration using the @samp{qXfer:btrace-conf:read}
42840 (@pxref{qXfer btrace-conf read}) packet.
42842 The configuration describes the branch trace format and configuration
42843 settings for that format. The following information is described:
42847 This thread uses the @dfn{Branch Trace Store} (@acronym{BTS}) format.
42850 The size of the @acronym{BTS} ring buffer in bytes.
42853 This thread uses the @dfn{Intel Processor Trace} (@acronym{Intel
42857 The size of the @acronym{Intel PT} ring buffer in bytes.
42861 @value{GDBN} must be linked with the Expat library to support XML
42862 branch trace configuration discovery. @xref{Expat}.
42864 The formal DTD for the branch trace configuration format is given below:
42867 <!ELEMENT btrace-conf (bts?, pt?)>
42868 <!ATTLIST btrace-conf version CDATA #FIXED "1.0">
42870 <!ELEMENT bts EMPTY>
42871 <!ATTLIST bts size CDATA #IMPLIED>
42873 <!ELEMENT pt EMPTY>
42874 <!ATTLIST pt size CDATA #IMPLIED>
42877 @include agentexpr.texi
42879 @node Target Descriptions
42880 @appendix Target Descriptions
42881 @cindex target descriptions
42883 One of the challenges of using @value{GDBN} to debug embedded systems
42884 is that there are so many minor variants of each processor
42885 architecture in use. It is common practice for vendors to start with
42886 a standard processor core --- ARM, PowerPC, or @acronym{MIPS}, for example ---
42887 and then make changes to adapt it to a particular market niche. Some
42888 architectures have hundreds of variants, available from dozens of
42889 vendors. This leads to a number of problems:
42893 With so many different customized processors, it is difficult for
42894 the @value{GDBN} maintainers to keep up with the changes.
42896 Since individual variants may have short lifetimes or limited
42897 audiences, it may not be worthwhile to carry information about every
42898 variant in the @value{GDBN} source tree.
42900 When @value{GDBN} does support the architecture of the embedded system
42901 at hand, the task of finding the correct architecture name to give the
42902 @command{set architecture} command can be error-prone.
42905 To address these problems, the @value{GDBN} remote protocol allows a
42906 target system to not only identify itself to @value{GDBN}, but to
42907 actually describe its own features. This lets @value{GDBN} support
42908 processor variants it has never seen before --- to the extent that the
42909 descriptions are accurate, and that @value{GDBN} understands them.
42911 @value{GDBN} must be linked with the Expat library to support XML
42912 target descriptions. @xref{Expat}.
42915 * Retrieving Descriptions:: How descriptions are fetched from a target.
42916 * Target Description Format:: The contents of a target description.
42917 * Predefined Target Types:: Standard types available for target
42919 * Enum Target Types:: How to define enum target types.
42920 * Standard Target Features:: Features @value{GDBN} knows about.
42923 @node Retrieving Descriptions
42924 @section Retrieving Descriptions
42926 Target descriptions can be read from the target automatically, or
42927 specified by the user manually. The default behavior is to read the
42928 description from the target. @value{GDBN} retrieves it via the remote
42929 protocol using @samp{qXfer} requests (@pxref{General Query Packets,
42930 qXfer}). The @var{annex} in the @samp{qXfer} packet will be
42931 @samp{target.xml}. The contents of the @samp{target.xml} annex are an
42932 XML document, of the form described in @ref{Target Description
42935 Alternatively, you can specify a file to read for the target description.
42936 If a file is set, the target will not be queried. The commands to
42937 specify a file are:
42940 @cindex set tdesc filename
42941 @item set tdesc filename @var{path}
42942 Read the target description from @var{path}.
42944 @cindex unset tdesc filename
42945 @item unset tdesc filename
42946 Do not read the XML target description from a file. @value{GDBN}
42947 will use the description supplied by the current target.
42949 @cindex show tdesc filename
42950 @item show tdesc filename
42951 Show the filename to read for a target description, if any.
42955 @node Target Description Format
42956 @section Target Description Format
42957 @cindex target descriptions, XML format
42959 A target description annex is an @uref{http://www.w3.org/XML/, XML}
42960 document which complies with the Document Type Definition provided in
42961 the @value{GDBN} sources in @file{gdb/features/gdb-target.dtd}. This
42962 means you can use generally available tools like @command{xmllint} to
42963 check that your feature descriptions are well-formed and valid.
42964 However, to help people unfamiliar with XML write descriptions for
42965 their targets, we also describe the grammar here.
42967 Target descriptions can identify the architecture of the remote target
42968 and (for some architectures) provide information about custom register
42969 sets. They can also identify the OS ABI of the remote target.
42970 @value{GDBN} can use this information to autoconfigure for your
42971 target, or to warn you if you connect to an unsupported target.
42973 Here is a simple target description:
42976 <target version="1.0">
42977 <architecture>i386:x86-64</architecture>
42982 This minimal description only says that the target uses
42983 the x86-64 architecture.
42985 A target description has the following overall form, with [ ] marking
42986 optional elements and @dots{} marking repeatable elements. The elements
42987 are explained further below.
42990 <?xml version="1.0"?>
42991 <!DOCTYPE target SYSTEM "gdb-target.dtd">
42992 <target version="1.0">
42993 @r{[}@var{architecture}@r{]}
42994 @r{[}@var{osabi}@r{]}
42995 @r{[}@var{compatible}@r{]}
42996 @r{[}@var{feature}@dots{}@r{]}
43001 The description is generally insensitive to whitespace and line
43002 breaks, under the usual common-sense rules. The XML version
43003 declaration and document type declaration can generally be omitted
43004 (@value{GDBN} does not require them), but specifying them may be
43005 useful for XML validation tools. The @samp{version} attribute for
43006 @samp{<target>} may also be omitted, but we recommend
43007 including it; if future versions of @value{GDBN} use an incompatible
43008 revision of @file{gdb-target.dtd}, they will detect and report
43009 the version mismatch.
43011 @subsection Inclusion
43012 @cindex target descriptions, inclusion
43015 @cindex <xi:include>
43018 It can sometimes be valuable to split a target description up into
43019 several different annexes, either for organizational purposes, or to
43020 share files between different possible target descriptions. You can
43021 divide a description into multiple files by replacing any element of
43022 the target description with an inclusion directive of the form:
43025 <xi:include href="@var{document}"/>
43029 When @value{GDBN} encounters an element of this form, it will retrieve
43030 the named XML @var{document}, and replace the inclusion directive with
43031 the contents of that document. If the current description was read
43032 using @samp{qXfer}, then so will be the included document;
43033 @var{document} will be interpreted as the name of an annex. If the
43034 current description was read from a file, @value{GDBN} will look for
43035 @var{document} as a file in the same directory where it found the
43036 original description.
43038 @subsection Architecture
43039 @cindex <architecture>
43041 An @samp{<architecture>} element has this form:
43044 <architecture>@var{arch}</architecture>
43047 @var{arch} is one of the architectures from the set accepted by
43048 @code{set architecture} (@pxref{Targets, ,Specifying a Debugging Target}).
43051 @cindex @code{<osabi>}
43053 This optional field was introduced in @value{GDBN} version 7.0.
43054 Previous versions of @value{GDBN} ignore it.
43056 An @samp{<osabi>} element has this form:
43059 <osabi>@var{abi-name}</osabi>
43062 @var{abi-name} is an OS ABI name from the same selection accepted by
43063 @w{@code{set osabi}} (@pxref{ABI, ,Configuring the Current ABI}).
43065 @subsection Compatible Architecture
43066 @cindex @code{<compatible>}
43068 This optional field was introduced in @value{GDBN} version 7.0.
43069 Previous versions of @value{GDBN} ignore it.
43071 A @samp{<compatible>} element has this form:
43074 <compatible>@var{arch}</compatible>
43077 @var{arch} is one of the architectures from the set accepted by
43078 @code{set architecture} (@pxref{Targets, ,Specifying a Debugging Target}).
43080 A @samp{<compatible>} element is used to specify that the target
43081 is able to run binaries in some other than the main target architecture
43082 given by the @samp{<architecture>} element. For example, on the
43083 Cell Broadband Engine, the main architecture is @code{powerpc:common}
43084 or @code{powerpc:common64}, but the system is able to run binaries
43085 in the @code{spu} architecture as well. The way to describe this
43086 capability with @samp{<compatible>} is as follows:
43089 <architecture>powerpc:common</architecture>
43090 <compatible>spu</compatible>
43093 @subsection Features
43096 Each @samp{<feature>} describes some logical portion of the target
43097 system. Features are currently used to describe available CPU
43098 registers and the types of their contents. A @samp{<feature>} element
43102 <feature name="@var{name}">
43103 @r{[}@var{type}@dots{}@r{]}
43109 Each feature's name should be unique within the description. The name
43110 of a feature does not matter unless @value{GDBN} has some special
43111 knowledge of the contents of that feature; if it does, the feature
43112 should have its standard name. @xref{Standard Target Features}.
43116 Any register's value is a collection of bits which @value{GDBN} must
43117 interpret. The default interpretation is a two's complement integer,
43118 but other types can be requested by name in the register description.
43119 Some predefined types are provided by @value{GDBN} (@pxref{Predefined
43120 Target Types}), and the description can define additional composite
43123 Each type element must have an @samp{id} attribute, which gives
43124 a unique (within the containing @samp{<feature>}) name to the type.
43125 Types must be defined before they are used.
43128 Some targets offer vector registers, which can be treated as arrays
43129 of scalar elements. These types are written as @samp{<vector>} elements,
43130 specifying the array element type, @var{type}, and the number of elements,
43134 <vector id="@var{id}" type="@var{type}" count="@var{count}"/>
43138 If a register's value is usefully viewed in multiple ways, define it
43139 with a union type containing the useful representations. The
43140 @samp{<union>} element contains one or more @samp{<field>} elements,
43141 each of which has a @var{name} and a @var{type}:
43144 <union id="@var{id}">
43145 <field name="@var{name}" type="@var{type}"/>
43152 If a register's value is composed from several separate values, define
43153 it with either a structure type or a flags type.
43154 A flags type may only contain bitfields.
43155 A structure type may either contain only bitfields or contain no bitfields.
43156 If the value contains only bitfields, its total size in bytes must be
43159 Non-bitfield values have a @var{name} and @var{type}.
43162 <struct id="@var{id}">
43163 <field name="@var{name}" type="@var{type}"/>
43168 Both @var{name} and @var{type} values are required.
43169 No implicit padding is added.
43171 Bitfield values have a @var{name}, @var{start}, @var{end} and @var{type}.
43174 <struct id="@var{id}" size="@var{size}">
43175 <field name="@var{name}" start="@var{start}" end="@var{end}" type="@var{type}"/>
43181 <flags id="@var{id}" size="@var{size}">
43182 <field name="@var{name}" start="@var{start}" end="@var{end}" type="@var{type}"/>
43187 The @var{name} value is required.
43188 Bitfield values may be named with the empty string, @samp{""},
43189 in which case the field is ``filler'' and its value is not printed.
43190 Not all bits need to be specified, so ``filler'' fields are optional.
43192 The @var{start} and @var{end} values are required, and @var{type}
43194 The field's @var{start} must be less than or equal to its @var{end},
43195 and zero represents the least significant bit.
43197 The default value of @var{type} is @code{bool} for single bit fields,
43198 and an unsigned integer otherwise.
43200 Which to choose? Structures or flags?
43202 Registers defined with @samp{flags} have these advantages over
43203 defining them with @samp{struct}:
43207 Arithmetic may be performed on them as if they were integers.
43209 They are printed in a more readable fashion.
43212 Registers defined with @samp{struct} have one advantage over
43213 defining them with @samp{flags}:
43217 One can fetch individual fields like in @samp{C}.
43220 (gdb) print $my_struct_reg.field3
43226 @subsection Registers
43229 Each register is represented as an element with this form:
43232 <reg name="@var{name}"
43233 bitsize="@var{size}"
43234 @r{[}regnum="@var{num}"@r{]}
43235 @r{[}save-restore="@var{save-restore}"@r{]}
43236 @r{[}type="@var{type}"@r{]}
43237 @r{[}group="@var{group}"@r{]}/>
43241 The components are as follows:
43246 The register's name; it must be unique within the target description.
43249 The register's size, in bits.
43252 The register's number. If omitted, a register's number is one greater
43253 than that of the previous register (either in the current feature or in
43254 a preceding feature); the first register in the target description
43255 defaults to zero. This register number is used to read or write
43256 the register; e.g.@: it is used in the remote @code{p} and @code{P}
43257 packets, and registers appear in the @code{g} and @code{G} packets
43258 in order of increasing register number.
43261 Whether the register should be preserved across inferior function
43262 calls; this must be either @code{yes} or @code{no}. The default is
43263 @code{yes}, which is appropriate for most registers except for
43264 some system control registers; this is not related to the target's
43268 The type of the register. It may be a predefined type, a type
43269 defined in the current feature, or one of the special types @code{int}
43270 and @code{float}. @code{int} is an integer type of the correct size
43271 for @var{bitsize}, and @code{float} is a floating point type (in the
43272 architecture's normal floating point format) of the correct size for
43273 @var{bitsize}. The default is @code{int}.
43276 The register group to which this register belongs. It can be one of the
43277 standard register groups @code{general}, @code{float}, @code{vector} or an
43278 arbitrary string. Group names should be limited to alphanumeric characters.
43279 If a group name is made up of multiple words the words may be separated by
43280 hyphens; e.g.@: @code{special-group} or @code{ultra-special-group}. If no
43281 @var{group} is specified, @value{GDBN} will not display the register in
43282 @code{info registers}.
43286 @node Predefined Target Types
43287 @section Predefined Target Types
43288 @cindex target descriptions, predefined types
43290 Type definitions in the self-description can build up composite types
43291 from basic building blocks, but can not define fundamental types. Instead,
43292 standard identifiers are provided by @value{GDBN} for the fundamental
43293 types. The currently supported types are:
43298 Boolean type, occupying a single bit.
43306 Signed integer types holding the specified number of bits.
43314 Unsigned integer types holding the specified number of bits.
43318 Pointers to unspecified code and data. The program counter and
43319 any dedicated return address register may be marked as code
43320 pointers; printing a code pointer converts it into a symbolic
43321 address. The stack pointer and any dedicated address registers
43322 may be marked as data pointers.
43325 Single precision IEEE floating point.
43328 Double precision IEEE floating point.
43331 The 12-byte extended precision format used by ARM FPA registers.
43334 The 10-byte extended precision format used by x87 registers.
43337 32bit @sc{eflags} register used by x86.
43340 32bit @sc{mxcsr} register used by x86.
43344 @node Enum Target Types
43345 @section Enum Target Types
43346 @cindex target descriptions, enum types
43348 Enum target types are useful in @samp{struct} and @samp{flags}
43349 register descriptions. @xref{Target Description Format}.
43351 Enum types have a name, size and a list of name/value pairs.
43354 <enum id="@var{id}" size="@var{size}">
43355 <evalue name="@var{name}" value="@var{value}"/>
43360 Enums must be defined before they are used.
43363 <enum id="levels_type" size="4">
43364 <evalue name="low" value="0"/>
43365 <evalue name="high" value="1"/>
43367 <flags id="flags_type" size="4">
43368 <field name="X" start="0"/>
43369 <field name="LEVEL" start="1" end="1" type="levels_type"/>
43371 <reg name="flags" bitsize="32" type="flags_type"/>
43374 Given that description, a value of 3 for the @samp{flags} register
43375 would be printed as:
43378 (gdb) info register flags
43379 flags 0x3 [ X LEVEL=high ]
43382 @node Standard Target Features
43383 @section Standard Target Features
43384 @cindex target descriptions, standard features
43386 A target description must contain either no registers or all the
43387 target's registers. If the description contains no registers, then
43388 @value{GDBN} will assume a default register layout, selected based on
43389 the architecture. If the description contains any registers, the
43390 default layout will not be used; the standard registers must be
43391 described in the target description, in such a way that @value{GDBN}
43392 can recognize them.
43394 This is accomplished by giving specific names to feature elements
43395 which contain standard registers. @value{GDBN} will look for features
43396 with those names and verify that they contain the expected registers;
43397 if any known feature is missing required registers, or if any required
43398 feature is missing, @value{GDBN} will reject the target
43399 description. You can add additional registers to any of the
43400 standard features --- @value{GDBN} will display them just as if
43401 they were added to an unrecognized feature.
43403 This section lists the known features and their expected contents.
43404 Sample XML documents for these features are included in the
43405 @value{GDBN} source tree, in the directory @file{gdb/features}.
43407 Names recognized by @value{GDBN} should include the name of the
43408 company or organization which selected the name, and the overall
43409 architecture to which the feature applies; so e.g.@: the feature
43410 containing ARM core registers is named @samp{org.gnu.gdb.arm.core}.
43412 The names of registers are not case sensitive for the purpose
43413 of recognizing standard features, but @value{GDBN} will only display
43414 registers using the capitalization used in the description.
43417 * AArch64 Features::
43421 * MicroBlaze Features::
43425 * Nios II Features::
43426 * OpenRISC 1000 Features::
43427 * PowerPC Features::
43428 * RISC-V Features::
43429 * S/390 and System z Features::
43435 @node AArch64 Features
43436 @subsection AArch64 Features
43437 @cindex target descriptions, AArch64 features
43439 The @samp{org.gnu.gdb.aarch64.core} feature is required for AArch64
43440 targets. It should contain registers @samp{x0} through @samp{x30},
43441 @samp{sp}, @samp{pc}, and @samp{cpsr}.
43443 The @samp{org.gnu.gdb.aarch64.fpu} feature is optional. If present,
43444 it should contain registers @samp{v0} through @samp{v31}, @samp{fpsr},
43447 The @samp{org.gnu.gdb.aarch64.sve} feature is optional. If present,
43448 it should contain registers @samp{z0} through @samp{z31}, @samp{p0}
43449 through @samp{p15}, @samp{ffr} and @samp{vg}.
43451 The @samp{org.gnu.gdb.aarch64.pauth} feature is optional. If present,
43452 it should contain registers @samp{pauth_dmask} and @samp{pauth_cmask}.
43455 @subsection ARC Features
43456 @cindex target descriptions, ARC Features
43458 ARC processors are highly configurable, so even core registers and their number
43459 are not completely predetermined. In addition flags and PC registers which are
43460 important to @value{GDBN} are not ``core'' registers in ARC. It is required
43461 that one of the core registers features is present.
43462 @samp{org.gnu.gdb.arc.aux-minimal} feature is mandatory.
43464 The @samp{org.gnu.gdb.arc.core.v2} feature is required for ARC EM and ARC HS
43465 targets with a normal register file. It should contain registers @samp{r0}
43466 through @samp{r25}, @samp{gp}, @samp{fp}, @samp{sp}, @samp{r30}, @samp{blink},
43467 @samp{lp_count} and @samp{pcl}. This feature may contain register @samp{ilink}
43468 and any of extension core registers @samp{r32} through @samp{r59/acch}.
43469 @samp{ilink} and extension core registers are not available to read/write, when
43470 debugging GNU/Linux applications, thus @samp{ilink} is made optional.
43472 The @samp{org.gnu.gdb.arc.core-reduced.v2} feature is required for ARC EM and
43473 ARC HS targets with a reduced register file. It should contain registers
43474 @samp{r0} through @samp{r3}, @samp{r10} through @samp{r15}, @samp{gp},
43475 @samp{fp}, @samp{sp}, @samp{r30}, @samp{blink}, @samp{lp_count} and @samp{pcl}.
43476 This feature may contain register @samp{ilink} and any of extension core
43477 registers @samp{r32} through @samp{r59/acch}.
43479 The @samp{org.gnu.gdb.arc.core.arcompact} feature is required for ARCompact
43480 targets with a normal register file. It should contain registers @samp{r0}
43481 through @samp{r25}, @samp{gp}, @samp{fp}, @samp{sp}, @samp{r30}, @samp{blink},
43482 @samp{lp_count} and @samp{pcl}. This feature may contain registers
43483 @samp{ilink1}, @samp{ilink2} and any of extension core registers @samp{r32}
43484 through @samp{r59/acch}. @samp{ilink1} and @samp{ilink2} and extension core
43485 registers are not available when debugging GNU/Linux applications. The only
43486 difference with @samp{org.gnu.gdb.arc.core.v2} feature is in the names of
43487 @samp{ilink1} and @samp{ilink2} registers and that @samp{r30} is mandatory in
43488 ARC v2, but @samp{ilink2} is optional on ARCompact.
43490 The @samp{org.gnu.gdb.arc.aux-minimal} feature is required for all ARC
43491 targets. It should contain registers @samp{pc} and @samp{status32}.
43494 @subsection ARM Features
43495 @cindex target descriptions, ARM features
43497 The @samp{org.gnu.gdb.arm.core} feature is required for non-M-profile
43499 It should contain registers @samp{r0} through @samp{r13}, @samp{sp},
43500 @samp{lr}, @samp{pc}, and @samp{cpsr}.
43502 For M-profile targets (e.g. Cortex-M3), the @samp{org.gnu.gdb.arm.core}
43503 feature is replaced by @samp{org.gnu.gdb.arm.m-profile}. It should contain
43504 registers @samp{r0} through @samp{r13}, @samp{sp}, @samp{lr}, @samp{pc},
43507 The @samp{org.gnu.gdb.arm.fpa} feature is optional. If present, it
43508 should contain registers @samp{f0} through @samp{f7} and @samp{fps}.
43510 The @samp{org.gnu.gdb.xscale.iwmmxt} feature is optional. If present,
43511 it should contain at least registers @samp{wR0} through @samp{wR15} and
43512 @samp{wCGR0} through @samp{wCGR3}. The @samp{wCID}, @samp{wCon},
43513 @samp{wCSSF}, and @samp{wCASF} registers are optional.
43515 The @samp{org.gnu.gdb.arm.vfp} feature is optional. If present, it
43516 should contain at least registers @samp{d0} through @samp{d15}. If
43517 they are present, @samp{d16} through @samp{d31} should also be included.
43518 @value{GDBN} will synthesize the single-precision registers from
43519 halves of the double-precision registers.
43521 The @samp{org.gnu.gdb.arm.neon} feature is optional. It does not
43522 need to contain registers; it instructs @value{GDBN} to display the
43523 VFP double-precision registers as vectors and to synthesize the
43524 quad-precision registers from pairs of double-precision registers.
43525 If this feature is present, @samp{org.gnu.gdb.arm.vfp} must also
43526 be present and include 32 double-precision registers.
43528 @node i386 Features
43529 @subsection i386 Features
43530 @cindex target descriptions, i386 features
43532 The @samp{org.gnu.gdb.i386.core} feature is required for i386/amd64
43533 targets. It should describe the following registers:
43537 @samp{eax} through @samp{edi} plus @samp{eip} for i386
43539 @samp{rax} through @samp{r15} plus @samp{rip} for amd64
43541 @samp{eflags}, @samp{cs}, @samp{ss}, @samp{ds}, @samp{es},
43542 @samp{fs}, @samp{gs}
43544 @samp{st0} through @samp{st7}
43546 @samp{fctrl}, @samp{fstat}, @samp{ftag}, @samp{fiseg}, @samp{fioff},
43547 @samp{foseg}, @samp{fooff} and @samp{fop}
43550 The register sets may be different, depending on the target.
43552 The @samp{org.gnu.gdb.i386.sse} feature is optional. It should
43553 describe registers:
43557 @samp{xmm0} through @samp{xmm7} for i386
43559 @samp{xmm0} through @samp{xmm15} for amd64
43564 The @samp{org.gnu.gdb.i386.avx} feature is optional and requires the
43565 @samp{org.gnu.gdb.i386.sse} feature. It should
43566 describe the upper 128 bits of @sc{ymm} registers:
43570 @samp{ymm0h} through @samp{ymm7h} for i386
43572 @samp{ymm0h} through @samp{ymm15h} for amd64
43575 The @samp{org.gnu.gdb.i386.mpx} is an optional feature representing Intel
43576 Memory Protection Extension (MPX). It should describe the following registers:
43580 @samp{bnd0raw} through @samp{bnd3raw} for i386 and amd64.
43582 @samp{bndcfgu} and @samp{bndstatus} for i386 and amd64.
43585 The @samp{org.gnu.gdb.i386.linux} feature is optional. It should
43586 describe a single register, @samp{orig_eax}.
43588 The @samp{org.gnu.gdb.i386.segments} feature is optional. It should
43589 describe two system registers: @samp{fs_base} and @samp{gs_base}.
43591 The @samp{org.gnu.gdb.i386.avx512} feature is optional and requires the
43592 @samp{org.gnu.gdb.i386.avx} feature. It should
43593 describe additional @sc{xmm} registers:
43597 @samp{xmm16h} through @samp{xmm31h}, only valid for amd64.
43600 It should describe the upper 128 bits of additional @sc{ymm} registers:
43604 @samp{ymm16h} through @samp{ymm31h}, only valid for amd64.
43608 describe the upper 256 bits of @sc{zmm} registers:
43612 @samp{zmm0h} through @samp{zmm7h} for i386.
43614 @samp{zmm0h} through @samp{zmm15h} for amd64.
43618 describe the additional @sc{zmm} registers:
43622 @samp{zmm16h} through @samp{zmm31h}, only valid for amd64.
43625 The @samp{org.gnu.gdb.i386.pkeys} feature is optional. It should
43626 describe a single register, @samp{pkru}. It is a 32-bit register
43627 valid for i386 and amd64.
43629 @node MicroBlaze Features
43630 @subsection MicroBlaze Features
43631 @cindex target descriptions, MicroBlaze features
43633 The @samp{org.gnu.gdb.microblaze.core} feature is required for MicroBlaze
43634 targets. It should contain registers @samp{r0} through @samp{r31},
43635 @samp{rpc}, @samp{rmsr}, @samp{rear}, @samp{resr}, @samp{rfsr}, @samp{rbtr},
43636 @samp{rpvr}, @samp{rpvr1} through @samp{rpvr11}, @samp{redr}, @samp{rpid},
43637 @samp{rzpr}, @samp{rtlbx}, @samp{rtlbsx}, @samp{rtlblo}, and @samp{rtlbhi}.
43639 The @samp{org.gnu.gdb.microblaze.stack-protect} feature is optional.
43640 If present, it should contain registers @samp{rshr} and @samp{rslr}
43642 @node MIPS Features
43643 @subsection @acronym{MIPS} Features
43644 @cindex target descriptions, @acronym{MIPS} features
43646 The @samp{org.gnu.gdb.mips.cpu} feature is required for @acronym{MIPS} targets.
43647 It should contain registers @samp{r0} through @samp{r31}, @samp{lo},
43648 @samp{hi}, and @samp{pc}. They may be 32-bit or 64-bit depending
43651 The @samp{org.gnu.gdb.mips.cp0} feature is also required. It should
43652 contain at least the @samp{status}, @samp{badvaddr}, and @samp{cause}
43653 registers. They may be 32-bit or 64-bit depending on the target.
43655 The @samp{org.gnu.gdb.mips.fpu} feature is currently required, though
43656 it may be optional in a future version of @value{GDBN}. It should
43657 contain registers @samp{f0} through @samp{f31}, @samp{fcsr}, and
43658 @samp{fir}. They may be 32-bit or 64-bit depending on the target.
43660 The @samp{org.gnu.gdb.mips.dsp} feature is optional. It should
43661 contain registers @samp{hi1} through @samp{hi3}, @samp{lo1} through
43662 @samp{lo3}, and @samp{dspctl}. The @samp{dspctl} register should
43663 be 32-bit and the rest may be 32-bit or 64-bit depending on the target.
43665 The @samp{org.gnu.gdb.mips.linux} feature is optional. It should
43666 contain a single register, @samp{restart}, which is used by the
43667 Linux kernel to control restartable syscalls.
43669 @node M68K Features
43670 @subsection M68K Features
43671 @cindex target descriptions, M68K features
43674 @item @samp{org.gnu.gdb.m68k.core}
43675 @itemx @samp{org.gnu.gdb.coldfire.core}
43676 @itemx @samp{org.gnu.gdb.fido.core}
43677 One of those features must be always present.
43678 The feature that is present determines which flavor of m68k is
43679 used. The feature that is present should contain registers
43680 @samp{d0} through @samp{d7}, @samp{a0} through @samp{a5}, @samp{fp},
43681 @samp{sp}, @samp{ps} and @samp{pc}.
43683 @item @samp{org.gnu.gdb.coldfire.fp}
43684 This feature is optional. If present, it should contain registers
43685 @samp{fp0} through @samp{fp7}, @samp{fpcontrol}, @samp{fpstatus} and
43689 @node NDS32 Features
43690 @subsection NDS32 Features
43691 @cindex target descriptions, NDS32 features
43693 The @samp{org.gnu.gdb.nds32.core} feature is required for NDS32
43694 targets. It should contain at least registers @samp{r0} through
43695 @samp{r10}, @samp{r15}, @samp{fp}, @samp{gp}, @samp{lp}, @samp{sp},
43698 The @samp{org.gnu.gdb.nds32.fpu} feature is optional. If present,
43699 it should contain 64-bit double-precision floating-point registers
43700 @samp{fd0} through @emph{fdN}, which should be @samp{fd3}, @samp{fd7},
43701 @samp{fd15}, or @samp{fd31} based on the FPU configuration implemented.
43703 @emph{Note:} The first sixteen 64-bit double-precision floating-point
43704 registers are overlapped with the thirty-two 32-bit single-precision
43705 floating-point registers. The 32-bit single-precision registers, if
43706 not being listed explicitly, will be synthesized from halves of the
43707 overlapping 64-bit double-precision registers. Listing 32-bit
43708 single-precision registers explicitly is deprecated, and the
43709 support to it could be totally removed some day.
43711 @node Nios II Features
43712 @subsection Nios II Features
43713 @cindex target descriptions, Nios II features
43715 The @samp{org.gnu.gdb.nios2.cpu} feature is required for Nios II
43716 targets. It should contain the 32 core registers (@samp{zero},
43717 @samp{at}, @samp{r2} through @samp{r23}, @samp{et} through @samp{ra}),
43718 @samp{pc}, and the 16 control registers (@samp{status} through
43721 @node OpenRISC 1000 Features
43722 @subsection Openrisc 1000 Features
43723 @cindex target descriptions, OpenRISC 1000 features
43725 The @samp{org.gnu.gdb.or1k.group0} feature is required for OpenRISC 1000
43726 targets. It should contain the 32 general purpose registers (@samp{r0}
43727 through @samp{r31}), @samp{ppc}, @samp{npc} and @samp{sr}.
43729 @node PowerPC Features
43730 @subsection PowerPC Features
43731 @cindex target descriptions, PowerPC features
43733 The @samp{org.gnu.gdb.power.core} feature is required for PowerPC
43734 targets. It should contain registers @samp{r0} through @samp{r31},
43735 @samp{pc}, @samp{msr}, @samp{cr}, @samp{lr}, @samp{ctr}, and
43736 @samp{xer}. They may be 32-bit or 64-bit depending on the target.
43738 The @samp{org.gnu.gdb.power.fpu} feature is optional. It should
43739 contain registers @samp{f0} through @samp{f31} and @samp{fpscr}.
43741 The @samp{org.gnu.gdb.power.altivec} feature is optional. It should
43742 contain registers @samp{vr0} through @samp{vr31}, @samp{vscr}, and
43743 @samp{vrsave}. @value{GDBN} will define pseudo-registers @samp{v0}
43744 through @samp{v31} as aliases for the corresponding @samp{vrX}
43747 The @samp{org.gnu.gdb.power.vsx} feature is optional. It should
43748 contain registers @samp{vs0h} through @samp{vs31h}. @value{GDBN} will
43749 combine these registers with the floating point registers (@samp{f0}
43750 through @samp{f31}) and the altivec registers (@samp{vr0} through
43751 @samp{vr31}) to present the 128-bit wide registers @samp{vs0} through
43752 @samp{vs63}, the set of vector-scalar registers for POWER7.
43753 Therefore, this feature requires both @samp{org.gnu.gdb.power.fpu} and
43754 @samp{org.gnu.gdb.power.altivec}.
43756 The @samp{org.gnu.gdb.power.spe} feature is optional. It should
43757 contain registers @samp{ev0h} through @samp{ev31h}, @samp{acc}, and
43758 @samp{spefscr}. SPE targets should provide 32-bit registers in
43759 @samp{org.gnu.gdb.power.core} and provide the upper halves in
43760 @samp{ev0h} through @samp{ev31h}. @value{GDBN} will combine
43761 these to present registers @samp{ev0} through @samp{ev31} to the
43764 The @samp{org.gnu.gdb.power.ppr} feature is optional. It should
43765 contain the 64-bit register @samp{ppr}.
43767 The @samp{org.gnu.gdb.power.dscr} feature is optional. It should
43768 contain the 64-bit register @samp{dscr}.
43770 The @samp{org.gnu.gdb.power.tar} feature is optional. It should
43771 contain the 64-bit register @samp{tar}.
43773 The @samp{org.gnu.gdb.power.ebb} feature is optional. It should
43774 contain registers @samp{bescr}, @samp{ebbhr} and @samp{ebbrr}, all
43777 The @samp{org.gnu.gdb.power.linux.pmu} feature is optional. It should
43778 contain registers @samp{mmcr0}, @samp{mmcr2}, @samp{siar}, @samp{sdar}
43779 and @samp{sier}, all 64-bit wide. This is the subset of the isa 2.07
43780 server PMU registers provided by @sc{gnu}/Linux.
43782 The @samp{org.gnu.gdb.power.htm.spr} feature is optional. It should
43783 contain registers @samp{tfhar}, @samp{texasr} and @samp{tfiar}, all
43786 The @samp{org.gnu.gdb.power.htm.core} feature is optional. It should
43787 contain the checkpointed general-purpose registers @samp{cr0} through
43788 @samp{cr31}, as well as the checkpointed registers @samp{clr} and
43789 @samp{cctr}. These registers may all be either 32-bit or 64-bit
43790 depending on the target. It should also contain the checkpointed
43791 registers @samp{ccr} and @samp{cxer}, which should both be 32-bit
43794 The @samp{org.gnu.gdb.power.htm.fpu} feature is optional. It should
43795 contain the checkpointed 64-bit floating-point registers @samp{cf0}
43796 through @samp{cf31}, as well as the checkpointed 64-bit register
43799 The @samp{org.gnu.gdb.power.htm.altivec} feature is optional. It
43800 should contain the checkpointed altivec registers @samp{cvr0} through
43801 @samp{cvr31}, all 128-bit wide. It should also contain the
43802 checkpointed registers @samp{cvscr} and @samp{cvrsave}, both 32-bit
43805 The @samp{org.gnu.gdb.power.htm.vsx} feature is optional. It should
43806 contain registers @samp{cvs0h} through @samp{cvs31h}. @value{GDBN}
43807 will combine these registers with the checkpointed floating point
43808 registers (@samp{cf0} through @samp{cf31}) and the checkpointed
43809 altivec registers (@samp{cvr0} through @samp{cvr31}) to present the
43810 128-bit wide checkpointed vector-scalar registers @samp{cvs0} through
43811 @samp{cvs63}. Therefore, this feature requires both
43812 @samp{org.gnu.gdb.power.htm.altivec} and
43813 @samp{org.gnu.gdb.power.htm.fpu}.
43815 The @samp{org.gnu.gdb.power.htm.ppr} feature is optional. It should
43816 contain the 64-bit checkpointed register @samp{cppr}.
43818 The @samp{org.gnu.gdb.power.htm.dscr} feature is optional. It should
43819 contain the 64-bit checkpointed register @samp{cdscr}.
43821 The @samp{org.gnu.gdb.power.htm.tar} feature is optional. It should
43822 contain the 64-bit checkpointed register @samp{ctar}.
43825 @node RISC-V Features
43826 @subsection RISC-V Features
43827 @cindex target descriptions, RISC-V Features
43829 The @samp{org.gnu.gdb.riscv.cpu} feature is required for RISC-V
43830 targets. It should contain the registers @samp{x0} through
43831 @samp{x31}, and @samp{pc}. Either the architectural names (@samp{x0},
43832 @samp{x1}, etc) can be used, or the ABI names (@samp{zero}, @samp{ra},
43835 The @samp{org.gnu.gdb.riscv.fpu} feature is optional. If present, it
43836 should contain registers @samp{f0} through @samp{f31}, @samp{fflags},
43837 @samp{frm}, and @samp{fcsr}. As with the cpu feature, either the
43838 architectural register names, or the ABI names can be used.
43840 The @samp{org.gnu.gdb.riscv.virtual} feature is optional. If present,
43841 it should contain registers that are not backed by real registers on
43842 the target, but are instead virtual, where the register value is
43843 derived from other target state. In many ways these are like
43844 @value{GDBN}s pseudo-registers, except implemented by the target.
43845 Currently the only register expected in this set is the one byte
43846 @samp{priv} register that contains the target's privilege level in the
43847 least significant two bits.
43849 The @samp{org.gnu.gdb.riscv.csr} feature is optional. If present, it
43850 should contain all of the target's standard CSRs. Standard CSRs are
43851 those defined in the RISC-V specification documents. There is some
43852 overlap between this feature and the fpu feature; the @samp{fflags},
43853 @samp{frm}, and @samp{fcsr} registers could be in either feature. The
43854 expectation is that these registers will be in the fpu feature if the
43855 target has floating point hardware, but can be moved into the csr
43856 feature if the target has the floating point control registers, but no
43857 other floating point hardware.
43859 @node S/390 and System z Features
43860 @subsection S/390 and System z Features
43861 @cindex target descriptions, S/390 features
43862 @cindex target descriptions, System z features
43864 The @samp{org.gnu.gdb.s390.core} feature is required for S/390 and
43865 System z targets. It should contain the PSW and the 16 general
43866 registers. In particular, System z targets should provide the 64-bit
43867 registers @samp{pswm}, @samp{pswa}, and @samp{r0} through @samp{r15}.
43868 S/390 targets should provide the 32-bit versions of these registers.
43869 A System z target that runs in 31-bit addressing mode should provide
43870 32-bit versions of @samp{pswm} and @samp{pswa}, as well as the general
43871 register's upper halves @samp{r0h} through @samp{r15h}, and their
43872 lower halves @samp{r0l} through @samp{r15l}.
43874 The @samp{org.gnu.gdb.s390.fpr} feature is required. It should
43875 contain the 64-bit registers @samp{f0} through @samp{f15}, and
43878 The @samp{org.gnu.gdb.s390.acr} feature is required. It should
43879 contain the 32-bit registers @samp{acr0} through @samp{acr15}.
43881 The @samp{org.gnu.gdb.s390.linux} feature is optional. It should
43882 contain the register @samp{orig_r2}, which is 64-bit wide on System z
43883 targets and 32-bit otherwise. In addition, the feature may contain
43884 the @samp{last_break} register, whose width depends on the addressing
43885 mode, as well as the @samp{system_call} register, which is always
43888 The @samp{org.gnu.gdb.s390.tdb} feature is optional. It should
43889 contain the 64-bit registers @samp{tdb0}, @samp{tac}, @samp{tct},
43890 @samp{atia}, and @samp{tr0} through @samp{tr15}.
43892 The @samp{org.gnu.gdb.s390.vx} feature is optional. It should contain
43893 64-bit wide registers @samp{v0l} through @samp{v15l}, which will be
43894 combined by @value{GDBN} with the floating point registers @samp{f0}
43895 through @samp{f15} to present the 128-bit wide vector registers
43896 @samp{v0} through @samp{v15}. In addition, this feature should
43897 contain the 128-bit wide vector registers @samp{v16} through
43900 The @samp{org.gnu.gdb.s390.gs} feature is optional. It should contain
43901 the 64-bit wide guarded-storage-control registers @samp{gsd},
43902 @samp{gssm}, and @samp{gsepla}.
43904 The @samp{org.gnu.gdb.s390.gsbc} feature is optional. It should contain
43905 the 64-bit wide guarded-storage broadcast control registers
43906 @samp{bc_gsd}, @samp{bc_gssm}, and @samp{bc_gsepla}.
43908 @node Sparc Features
43909 @subsection Sparc Features
43910 @cindex target descriptions, sparc32 features
43911 @cindex target descriptions, sparc64 features
43912 The @samp{org.gnu.gdb.sparc.cpu} feature is required for sparc32/sparc64
43913 targets. It should describe the following registers:
43917 @samp{g0} through @samp{g7}
43919 @samp{o0} through @samp{o7}
43921 @samp{l0} through @samp{l7}
43923 @samp{i0} through @samp{i7}
43926 They may be 32-bit or 64-bit depending on the target.
43928 Also the @samp{org.gnu.gdb.sparc.fpu} feature is required for sparc32/sparc64
43929 targets. It should describe the following registers:
43933 @samp{f0} through @samp{f31}
43935 @samp{f32} through @samp{f62} for sparc64
43938 The @samp{org.gnu.gdb.sparc.cp0} feature is required for sparc32/sparc64
43939 targets. It should describe the following registers:
43943 @samp{y}, @samp{psr}, @samp{wim}, @samp{tbr}, @samp{pc}, @samp{npc},
43944 @samp{fsr}, and @samp{csr} for sparc32
43946 @samp{pc}, @samp{npc}, @samp{state}, @samp{fsr}, @samp{fprs}, and @samp{y}
43950 @node TIC6x Features
43951 @subsection TMS320C6x Features
43952 @cindex target descriptions, TIC6x features
43953 @cindex target descriptions, TMS320C6x features
43954 The @samp{org.gnu.gdb.tic6x.core} feature is required for TMS320C6x
43955 targets. It should contain registers @samp{A0} through @samp{A15},
43956 registers @samp{B0} through @samp{B15}, @samp{CSR} and @samp{PC}.
43958 The @samp{org.gnu.gdb.tic6x.gp} feature is optional. It should
43959 contain registers @samp{A16} through @samp{A31} and @samp{B16}
43960 through @samp{B31}.
43962 The @samp{org.gnu.gdb.tic6x.c6xp} feature is optional. It should
43963 contain registers @samp{TSR}, @samp{ILC} and @samp{RILC}.
43965 @node Operating System Information
43966 @appendix Operating System Information
43967 @cindex operating system information
43973 Users of @value{GDBN} often wish to obtain information about the state of
43974 the operating system running on the target---for example the list of
43975 processes, or the list of open files. This section describes the
43976 mechanism that makes it possible. This mechanism is similar to the
43977 target features mechanism (@pxref{Target Descriptions}), but focuses
43978 on a different aspect of target.
43980 Operating system information is retrived from the target via the
43981 remote protocol, using @samp{qXfer} requests (@pxref{qXfer osdata
43982 read}). The object name in the request should be @samp{osdata}, and
43983 the @var{annex} identifies the data to be fetched.
43986 @appendixsection Process list
43987 @cindex operating system information, process list
43989 When requesting the process list, the @var{annex} field in the
43990 @samp{qXfer} request should be @samp{processes}. The returned data is
43991 an XML document. The formal syntax of this document is defined in
43992 @file{gdb/features/osdata.dtd}.
43994 An example document is:
43997 <?xml version="1.0"?>
43998 <!DOCTYPE target SYSTEM "osdata.dtd">
43999 <osdata type="processes">
44001 <column name="pid">1</column>
44002 <column name="user">root</column>
44003 <column name="command">/sbin/init</column>
44004 <column name="cores">1,2,3</column>
44009 Each item should include a column whose name is @samp{pid}. The value
44010 of that column should identify the process on the target. The
44011 @samp{user} and @samp{command} columns are optional, and will be
44012 displayed by @value{GDBN}. The @samp{cores} column, if present,
44013 should contain a comma-separated list of cores that this process
44014 is running on. Target may provide additional columns,
44015 which @value{GDBN} currently ignores.
44017 @node Trace File Format
44018 @appendix Trace File Format
44019 @cindex trace file format
44021 The trace file comes in three parts: a header, a textual description
44022 section, and a trace frame section with binary data.
44024 The header has the form @code{\x7fTRACE0\n}. The first byte is
44025 @code{0x7f} so as to indicate that the file contains binary data,
44026 while the @code{0} is a version number that may have different values
44029 The description section consists of multiple lines of @sc{ascii} text
44030 separated by newline characters (@code{0xa}). The lines may include a
44031 variety of optional descriptive or context-setting information, such
44032 as tracepoint definitions or register set size. @value{GDBN} will
44033 ignore any line that it does not recognize. An empty line marks the end
44038 Specifies the size of a register block in bytes. This is equal to the
44039 size of a @code{g} packet payload in the remote protocol. @var{size}
44040 is an ascii decimal number. There should be only one such line in
44041 a single trace file.
44043 @item status @var{status}
44044 Trace status. @var{status} has the same format as a @code{qTStatus}
44045 remote packet reply. There should be only one such line in a single trace
44048 @item tp @var{payload}
44049 Tracepoint definition. The @var{payload} has the same format as
44050 @code{qTfP}/@code{qTsP} remote packet reply payload. A single tracepoint
44051 may take multiple lines of definition, corresponding to the multiple
44054 @item tsv @var{payload}
44055 Trace state variable definition. The @var{payload} has the same format as
44056 @code{qTfV}/@code{qTsV} remote packet reply payload. A single variable
44057 may take multiple lines of definition, corresponding to the multiple
44060 @item tdesc @var{payload}
44061 Target description in XML format. The @var{payload} is a single line of
44062 the XML file. All such lines should be concatenated together to get
44063 the original XML file. This file is in the same format as @code{qXfer}
44064 @code{features} payload, and corresponds to the main @code{target.xml}
44065 file. Includes are not allowed.
44069 The trace frame section consists of a number of consecutive frames.
44070 Each frame begins with a two-byte tracepoint number, followed by a
44071 four-byte size giving the amount of data in the frame. The data in
44072 the frame consists of a number of blocks, each introduced by a
44073 character indicating its type (at least register, memory, and trace
44074 state variable). The data in this section is raw binary, not a
44075 hexadecimal or other encoding; its endianness matches the target's
44078 @c FIXME bi-arch may require endianness/arch info in description section
44081 @item R @var{bytes}
44082 Register block. The number and ordering of bytes matches that of a
44083 @code{g} packet in the remote protocol. Note that these are the
44084 actual bytes, in target order, not a hexadecimal encoding.
44086 @item M @var{address} @var{length} @var{bytes}...
44087 Memory block. This is a contiguous block of memory, at the 8-byte
44088 address @var{address}, with a 2-byte length @var{length}, followed by
44089 @var{length} bytes.
44091 @item V @var{number} @var{value}
44092 Trace state variable block. This records the 8-byte signed value
44093 @var{value} of trace state variable numbered @var{number}.
44097 Future enhancements of the trace file format may include additional types
44100 @node Index Section Format
44101 @appendix @code{.gdb_index} section format
44102 @cindex .gdb_index section format
44103 @cindex index section format
44105 This section documents the index section that is created by @code{save
44106 gdb-index} (@pxref{Index Files}). The index section is
44107 DWARF-specific; some knowledge of DWARF is assumed in this
44110 The mapped index file format is designed to be directly
44111 @code{mmap}able on any architecture. In most cases, a datum is
44112 represented using a little-endian 32-bit integer value, called an
44113 @code{offset_type}. Big endian machines must byte-swap the values
44114 before using them. Exceptions to this rule are noted. The data is
44115 laid out such that alignment is always respected.
44117 A mapped index consists of several areas, laid out in order.
44121 The file header. This is a sequence of values, of @code{offset_type}
44122 unless otherwise noted:
44126 The version number, currently 8. Versions 1, 2 and 3 are obsolete.
44127 Version 4 uses a different hashing function from versions 5 and 6.
44128 Version 6 includes symbols for inlined functions, whereas versions 4
44129 and 5 do not. Version 7 adds attributes to the CU indices in the
44130 symbol table. Version 8 specifies that symbols from DWARF type units
44131 (@samp{DW_TAG_type_unit}) refer to the type unit's symbol table and not the
44132 compilation unit (@samp{DW_TAG_comp_unit}) using the type.
44134 @value{GDBN} will only read version 4, 5, or 6 indices
44135 by specifying @code{set use-deprecated-index-sections on}.
44136 GDB has a workaround for potentially broken version 7 indices so it is
44137 currently not flagged as deprecated.
44140 The offset, from the start of the file, of the CU list.
44143 The offset, from the start of the file, of the types CU list. Note
44144 that this area can be empty, in which case this offset will be equal
44145 to the next offset.
44148 The offset, from the start of the file, of the address area.
44151 The offset, from the start of the file, of the symbol table.
44154 The offset, from the start of the file, of the constant pool.
44158 The CU list. This is a sequence of pairs of 64-bit little-endian
44159 values, sorted by the CU offset. The first element in each pair is
44160 the offset of a CU in the @code{.debug_info} section. The second
44161 element in each pair is the length of that CU. References to a CU
44162 elsewhere in the map are done using a CU index, which is just the
44163 0-based index into this table. Note that if there are type CUs, then
44164 conceptually CUs and type CUs form a single list for the purposes of
44168 The types CU list. This is a sequence of triplets of 64-bit
44169 little-endian values. In a triplet, the first value is the CU offset,
44170 the second value is the type offset in the CU, and the third value is
44171 the type signature. The types CU list is not sorted.
44174 The address area. The address area consists of a sequence of address
44175 entries. Each address entry has three elements:
44179 The low address. This is a 64-bit little-endian value.
44182 The high address. This is a 64-bit little-endian value. Like
44183 @code{DW_AT_high_pc}, the value is one byte beyond the end.
44186 The CU index. This is an @code{offset_type} value.
44190 The symbol table. This is an open-addressed hash table. The size of
44191 the hash table is always a power of 2.
44193 Each slot in the hash table consists of a pair of @code{offset_type}
44194 values. The first value is the offset of the symbol's name in the
44195 constant pool. The second value is the offset of the CU vector in the
44198 If both values are 0, then this slot in the hash table is empty. This
44199 is ok because while 0 is a valid constant pool index, it cannot be a
44200 valid index for both a string and a CU vector.
44202 The hash value for a table entry is computed by applying an
44203 iterative hash function to the symbol's name. Starting with an
44204 initial value of @code{r = 0}, each (unsigned) character @samp{c} in
44205 the string is incorporated into the hash using the formula depending on the
44210 The formula is @code{r = r * 67 + c - 113}.
44212 @item Versions 5 to 7
44213 The formula is @code{r = r * 67 + tolower (c) - 113}.
44216 The terminating @samp{\0} is not incorporated into the hash.
44218 The step size used in the hash table is computed via
44219 @code{((hash * 17) & (size - 1)) | 1}, where @samp{hash} is the hash
44220 value, and @samp{size} is the size of the hash table. The step size
44221 is used to find the next candidate slot when handling a hash
44224 The names of C@t{++} symbols in the hash table are canonicalized. We
44225 don't currently have a simple description of the canonicalization
44226 algorithm; if you intend to create new index sections, you must read
44230 The constant pool. This is simply a bunch of bytes. It is organized
44231 so that alignment is correct: CU vectors are stored first, followed by
44234 A CU vector in the constant pool is a sequence of @code{offset_type}
44235 values. The first value is the number of CU indices in the vector.
44236 Each subsequent value is the index and symbol attributes of a CU in
44237 the CU list. This element in the hash table is used to indicate which
44238 CUs define the symbol and how the symbol is used.
44239 See below for the format of each CU index+attributes entry.
44241 A string in the constant pool is zero-terminated.
44244 Attributes were added to CU index values in @code{.gdb_index} version 7.
44245 If a symbol has multiple uses within a CU then there is one
44246 CU index+attributes value for each use.
44248 The format of each CU index+attributes entry is as follows
44254 This is the index of the CU in the CU list.
44256 These bits are reserved for future purposes and must be zero.
44258 The kind of the symbol in the CU.
44262 This value is reserved and should not be used.
44263 By reserving zero the full @code{offset_type} value is backwards compatible
44264 with previous versions of the index.
44266 The symbol is a type.
44268 The symbol is a variable or an enum value.
44270 The symbol is a function.
44272 Any other kind of symbol.
44274 These values are reserved.
44278 This bit is zero if the value is global and one if it is static.
44280 The determination of whether a symbol is global or static is complicated.
44281 The authorative reference is the file @file{dwarf2read.c} in
44282 @value{GDBN} sources.
44286 This pseudo-code describes the computation of a symbol's kind and
44287 global/static attributes in the index.
44290 is_external = get_attribute (die, DW_AT_external);
44291 language = get_attribute (cu_die, DW_AT_language);
44294 case DW_TAG_typedef:
44295 case DW_TAG_base_type:
44296 case DW_TAG_subrange_type:
44300 case DW_TAG_enumerator:
44302 is_static = language != CPLUS;
44304 case DW_TAG_subprogram:
44306 is_static = ! (is_external || language == ADA);
44308 case DW_TAG_constant:
44310 is_static = ! is_external;
44312 case DW_TAG_variable:
44314 is_static = ! is_external;
44316 case DW_TAG_namespace:
44320 case DW_TAG_class_type:
44321 case DW_TAG_interface_type:
44322 case DW_TAG_structure_type:
44323 case DW_TAG_union_type:
44324 case DW_TAG_enumeration_type:
44326 is_static = language != CPLUS;
44334 @appendix Manual pages
44338 * gdb man:: The GNU Debugger man page
44339 * gdbserver man:: Remote Server for the GNU Debugger man page
44340 * gcore man:: Generate a core file of a running program
44341 * gdbinit man:: gdbinit scripts
44342 * gdb-add-index man:: Add index files to speed up GDB
44348 @c man title gdb The GNU Debugger
44350 @c man begin SYNOPSIS gdb
44351 gdb [@option{-help}] [@option{-nh}] [@option{-nx}] [@option{-q}]
44352 [@option{-batch}] [@option{-cd=}@var{dir}] [@option{-f}]
44353 [@option{-b}@w{ }@var{bps}]
44354 [@option{-tty=}@var{dev}] [@option{-s} @var{symfile}]
44355 [@option{-e}@w{ }@var{prog}] [@option{-se}@w{ }@var{prog}]
44356 [@option{-c}@w{ }@var{core}] [@option{-p}@w{ }@var{procID}]
44357 [@option{-x}@w{ }@var{cmds}] [@option{-d}@w{ }@var{dir}]
44358 [@var{prog}|@var{prog} @var{procID}|@var{prog} @var{core}]
44361 @c man begin DESCRIPTION gdb
44362 The purpose of a debugger such as @value{GDBN} is to allow you to see what is
44363 going on ``inside'' another program while it executes -- or what another
44364 program was doing at the moment it crashed.
44366 @value{GDBN} can do four main kinds of things (plus other things in support of
44367 these) to help you catch bugs in the act:
44371 Start your program, specifying anything that might affect its behavior.
44374 Make your program stop on specified conditions.
44377 Examine what has happened, when your program has stopped.
44380 Change things in your program, so you can experiment with correcting the
44381 effects of one bug and go on to learn about another.
44384 You can use @value{GDBN} to debug programs written in C, C@t{++}, Fortran and
44387 @value{GDBN} is invoked with the shell command @code{gdb}. Once started, it reads
44388 commands from the terminal until you tell it to exit with the @value{GDBN}
44389 command @code{quit}. You can get online help from @value{GDBN} itself
44390 by using the command @code{help}.
44392 You can run @code{gdb} with no arguments or options; but the most
44393 usual way to start @value{GDBN} is with one argument or two, specifying an
44394 executable program as the argument:
44400 You can also start with both an executable program and a core file specified:
44406 You can, instead, specify a process ID as a second argument, if you want
44407 to debug a running process:
44415 would attach @value{GDBN} to process @code{1234} (unless you also have a file
44416 named @file{1234}; @value{GDBN} does check for a core file first).
44417 With option @option{-p} you can omit the @var{program} filename.
44419 Here are some of the most frequently needed @value{GDBN} commands:
44421 @c pod2man highlights the right hand side of the @item lines.
44423 @item break [@var{file}:]@var{function}
44424 Set a breakpoint at @var{function} (in @var{file}).
44426 @item run [@var{arglist}]
44427 Start your program (with @var{arglist}, if specified).
44430 Backtrace: display the program stack.
44432 @item print @var{expr}
44433 Display the value of an expression.
44436 Continue running your program (after stopping, e.g. at a breakpoint).
44439 Execute next program line (after stopping); step @emph{over} any
44440 function calls in the line.
44442 @item edit [@var{file}:]@var{function}
44443 look at the program line where it is presently stopped.
44445 @item list [@var{file}:]@var{function}
44446 type the text of the program in the vicinity of where it is presently stopped.
44449 Execute next program line (after stopping); step @emph{into} any
44450 function calls in the line.
44452 @item help [@var{name}]
44453 Show information about @value{GDBN} command @var{name}, or general information
44454 about using @value{GDBN}.
44457 Exit from @value{GDBN}.
44461 For full details on @value{GDBN},
44462 see @cite{Using GDB: A Guide to the GNU Source-Level Debugger},
44463 by Richard M. Stallman and Roland H. Pesch. The same text is available online
44464 as the @code{gdb} entry in the @code{info} program.
44468 @c man begin OPTIONS gdb
44469 Any arguments other than options specify an executable
44470 file and core file (or process ID); that is, the first argument
44471 encountered with no
44472 associated option flag is equivalent to a @option{-se} option, and the second,
44473 if any, is equivalent to a @option{-c} option if it's the name of a file.
44475 both long and short forms; both are shown here. The long forms are also
44476 recognized if you truncate them, so long as enough of the option is
44477 present to be unambiguous. (If you prefer, you can flag option
44478 arguments with @option{+} rather than @option{-}, though we illustrate the
44479 more usual convention.)
44481 All the options and command line arguments you give are processed
44482 in sequential order. The order makes a difference when the @option{-x}
44488 List all options, with brief explanations.
44490 @item -symbols=@var{file}
44491 @itemx -s @var{file}
44492 Read symbol table from file @var{file}.
44495 Enable writing into executable and core files.
44497 @item -exec=@var{file}
44498 @itemx -e @var{file}
44499 Use file @var{file} as the executable file to execute when
44500 appropriate, and for examining pure data in conjunction with a core
44503 @item -se=@var{file}
44504 Read symbol table from file @var{file} and use it as the executable
44507 @item -core=@var{file}
44508 @itemx -c @var{file}
44509 Use file @var{file} as a core dump to examine.
44511 @item -command=@var{file}
44512 @itemx -x @var{file}
44513 Execute @value{GDBN} commands from file @var{file}.
44515 @item -ex @var{command}
44516 Execute given @value{GDBN} @var{command}.
44518 @item -directory=@var{directory}
44519 @itemx -d @var{directory}
44520 Add @var{directory} to the path to search for source files.
44523 Do not execute commands from @file{~/.gdbinit}.
44527 Do not execute commands from any @file{.gdbinit} initialization files.
44531 ``Quiet''. Do not print the introductory and copyright messages. These
44532 messages are also suppressed in batch mode.
44535 Run in batch mode. Exit with status @code{0} after processing all the command
44536 files specified with @option{-x} (and @file{.gdbinit}, if not inhibited).
44537 Exit with nonzero status if an error occurs in executing the @value{GDBN}
44538 commands in the command files.
44540 Batch mode may be useful for running @value{GDBN} as a filter, for example to
44541 download and run a program on another computer; in order to make this
44542 more useful, the message
44545 Program exited normally.
44549 (which is ordinarily issued whenever a program running under @value{GDBN} control
44550 terminates) is not issued when running in batch mode.
44552 @item -cd=@var{directory}
44553 Run @value{GDBN} using @var{directory} as its working directory,
44554 instead of the current directory.
44558 Emacs sets this option when it runs @value{GDBN} as a subprocess. It tells
44559 @value{GDBN} to output the full file name and line number in a standard,
44560 recognizable fashion each time a stack frame is displayed (which
44561 includes each time the program stops). This recognizable format looks
44562 like two @samp{\032} characters, followed by the file name, line number
44563 and character position separated by colons, and a newline. The
44564 Emacs-to-@value{GDBN} interface program uses the two @samp{\032}
44565 characters as a signal to display the source code for the frame.
44568 Set the line speed (baud rate or bits per second) of any serial
44569 interface used by @value{GDBN} for remote debugging.
44571 @item -tty=@var{device}
44572 Run using @var{device} for your program's standard input and output.
44576 @c man begin SEEALSO gdb
44578 The full documentation for @value{GDBN} is maintained as a Texinfo manual.
44579 If the @code{info} and @code{gdb} programs and @value{GDBN}'s Texinfo
44580 documentation are properly installed at your site, the command
44587 should give you access to the complete manual.
44589 @cite{Using GDB: A Guide to the GNU Source-Level Debugger},
44590 Richard M. Stallman and Roland H. Pesch, July 1991.
44594 @node gdbserver man
44595 @heading gdbserver man
44597 @c man title gdbserver Remote Server for the GNU Debugger
44599 @c man begin SYNOPSIS gdbserver
44600 gdbserver @var{comm} @var{prog} [@var{args}@dots{}]
44602 gdbserver --attach @var{comm} @var{pid}
44604 gdbserver --multi @var{comm}
44608 @c man begin DESCRIPTION gdbserver
44609 @command{gdbserver} is a program that allows you to run @value{GDBN} on a different machine
44610 than the one which is running the program being debugged.
44613 @subheading Usage (server (target) side)
44616 Usage (server (target) side):
44619 First, you need to have a copy of the program you want to debug put onto
44620 the target system. The program can be stripped to save space if needed, as
44621 @command{gdbserver} doesn't care about symbols. All symbol handling is taken care of by
44622 the @value{GDBN} running on the host system.
44624 To use the server, you log on to the target system, and run the @command{gdbserver}
44625 program. You must tell it (a) how to communicate with @value{GDBN}, (b) the name of
44626 your program, and (c) its arguments. The general syntax is:
44629 target> gdbserver @var{comm} @var{program} [@var{args} ...]
44632 For example, using a serial port, you might say:
44636 @c @file would wrap it as F</dev/com1>.
44637 target> gdbserver /dev/com1 emacs foo.txt
44640 target> gdbserver @file{/dev/com1} emacs foo.txt
44644 This tells @command{gdbserver} to debug emacs with an argument of foo.txt, and
44645 to communicate with @value{GDBN} via @file{/dev/com1}. @command{gdbserver} now
44646 waits patiently for the host @value{GDBN} to communicate with it.
44648 To use a TCP connection, you could say:
44651 target> gdbserver host:2345 emacs foo.txt
44654 This says pretty much the same thing as the last example, except that we are
44655 going to communicate with the @code{host} @value{GDBN} via TCP. The @code{host:2345} argument means
44656 that we are expecting to see a TCP connection from @code{host} to local TCP port
44657 2345. (Currently, the @code{host} part is ignored.) You can choose any number you
44658 want for the port number as long as it does not conflict with any existing TCP
44659 ports on the target system. This same port number must be used in the host
44660 @value{GDBN}s @code{target remote} command, which will be described shortly. Note that if
44661 you chose a port number that conflicts with another service, @command{gdbserver} will
44662 print an error message and exit.
44664 @command{gdbserver} can also attach to running programs.
44665 This is accomplished via the @option{--attach} argument. The syntax is:
44668 target> gdbserver --attach @var{comm} @var{pid}
44671 @var{pid} is the process ID of a currently running process. It isn't
44672 necessary to point @command{gdbserver} at a binary for the running process.
44674 To start @code{gdbserver} without supplying an initial command to run
44675 or process ID to attach, use the @option{--multi} command line option.
44676 In such case you should connect using @kbd{target extended-remote} to start
44677 the program you want to debug.
44680 target> gdbserver --multi @var{comm}
44684 @subheading Usage (host side)
44690 You need an unstripped copy of the target program on your host system, since
44691 @value{GDBN} needs to examine its symbol tables and such. Start up @value{GDBN} as you normally
44692 would, with the target program as the first argument. (You may need to use the
44693 @option{--baud} option if the serial line is running at anything except 9600 baud.)
44694 That is @code{gdb TARGET-PROG}, or @code{gdb --baud BAUD TARGET-PROG}. After that, the only
44695 new command you need to know about is @code{target remote}
44696 (or @code{target extended-remote}). Its argument is either
44697 a device name (usually a serial device, like @file{/dev/ttyb}), or a @code{HOST:PORT}
44698 descriptor. For example:
44702 @c @file would wrap it as F</dev/ttyb>.
44703 (gdb) target remote /dev/ttyb
44706 (gdb) target remote @file{/dev/ttyb}
44711 communicates with the server via serial line @file{/dev/ttyb}, and:
44714 (gdb) target remote the-target:2345
44718 communicates via a TCP connection to port 2345 on host `the-target', where
44719 you previously started up @command{gdbserver} with the same port number. Note that for
44720 TCP connections, you must start up @command{gdbserver} prior to using the `target remote'
44721 command, otherwise you may get an error that looks something like
44722 `Connection refused'.
44724 @command{gdbserver} can also debug multiple inferiors at once,
44727 the @value{GDBN} manual in node @code{Inferiors and Programs}
44728 -- shell command @code{info -f gdb -n 'Inferiors and Programs'}.
44731 @ref{Inferiors and Programs}.
44733 In such case use the @code{extended-remote} @value{GDBN} command variant:
44736 (gdb) target extended-remote the-target:2345
44739 The @command{gdbserver} option @option{--multi} may or may not be used in such
44743 @c man begin OPTIONS gdbserver
44744 There are three different modes for invoking @command{gdbserver}:
44749 Debug a specific program specified by its program name:
44752 gdbserver @var{comm} @var{prog} [@var{args}@dots{}]
44755 The @var{comm} parameter specifies how should the server communicate
44756 with @value{GDBN}; it is either a device name (to use a serial line),
44757 a TCP port number (@code{:1234}), or @code{-} or @code{stdio} to use
44758 stdin/stdout of @code{gdbserver}. Specify the name of the program to
44759 debug in @var{prog}. Any remaining arguments will be passed to the
44760 program verbatim. When the program exits, @value{GDBN} will close the
44761 connection, and @code{gdbserver} will exit.
44764 Debug a specific program by specifying the process ID of a running
44768 gdbserver --attach @var{comm} @var{pid}
44771 The @var{comm} parameter is as described above. Supply the process ID
44772 of a running program in @var{pid}; @value{GDBN} will do everything
44773 else. Like with the previous mode, when the process @var{pid} exits,
44774 @value{GDBN} will close the connection, and @code{gdbserver} will exit.
44777 Multi-process mode -- debug more than one program/process:
44780 gdbserver --multi @var{comm}
44783 In this mode, @value{GDBN} can instruct @command{gdbserver} which
44784 command(s) to run. Unlike the other 2 modes, @value{GDBN} will not
44785 close the connection when a process being debugged exits, so you can
44786 debug several processes in the same session.
44789 In each of the modes you may specify these options:
44794 List all options, with brief explanations.
44797 This option causes @command{gdbserver} to print its version number and exit.
44800 @command{gdbserver} will attach to a running program. The syntax is:
44803 target> gdbserver --attach @var{comm} @var{pid}
44806 @var{pid} is the process ID of a currently running process. It isn't
44807 necessary to point @command{gdbserver} at a binary for the running process.
44810 To start @code{gdbserver} without supplying an initial command to run
44811 or process ID to attach, use this command line option.
44812 Then you can connect using @kbd{target extended-remote} and start
44813 the program you want to debug. The syntax is:
44816 target> gdbserver --multi @var{comm}
44820 Instruct @code{gdbserver} to display extra status information about the debugging
44822 This option is intended for @code{gdbserver} development and for bug reports to
44825 @item --remote-debug
44826 Instruct @code{gdbserver} to display remote protocol debug output.
44827 This option is intended for @code{gdbserver} development and for bug reports to
44830 @item --debug-file=@var{filename}
44831 Instruct @code{gdbserver} to send any debug output to the given @var{filename}.
44832 This option is intended for @code{gdbserver} development and for bug reports to
44835 @item --debug-format=option1@r{[},option2,...@r{]}
44836 Instruct @code{gdbserver} to include extra information in each line
44837 of debugging output.
44838 @xref{Other Command-Line Arguments for gdbserver}.
44841 Specify a wrapper to launch programs
44842 for debugging. The option should be followed by the name of the
44843 wrapper, then any command-line arguments to pass to the wrapper, then
44844 @kbd{--} indicating the end of the wrapper arguments.
44847 By default, @command{gdbserver} keeps the listening TCP port open, so that
44848 additional connections are possible. However, if you start @code{gdbserver}
44849 with the @option{--once} option, it will stop listening for any further
44850 connection attempts after connecting to the first @value{GDBN} session.
44852 @c --disable-packet is not documented for users.
44854 @c --disable-randomization and --no-disable-randomization are superseded by
44855 @c QDisableRandomization.
44860 @c man begin SEEALSO gdbserver
44862 The full documentation for @value{GDBN} is maintained as a Texinfo manual.
44863 If the @code{info} and @code{gdb} programs and @value{GDBN}'s Texinfo
44864 documentation are properly installed at your site, the command
44870 should give you access to the complete manual.
44872 @cite{Using GDB: A Guide to the GNU Source-Level Debugger},
44873 Richard M. Stallman and Roland H. Pesch, July 1991.
44880 @c man title gcore Generate a core file of a running program
44883 @c man begin SYNOPSIS gcore
44884 gcore [-a] [-o @var{prefix}] @var{pid1} [@var{pid2}...@var{pidN}]
44888 @c man begin DESCRIPTION gcore
44889 Generate core dumps of one or more running programs with process IDs
44890 @var{pid1}, @var{pid2}, etc. A core file produced by @command{gcore}
44891 is equivalent to one produced by the kernel when the process crashes
44892 (and when @kbd{ulimit -c} was used to set up an appropriate core dump
44893 limit). However, unlike after a crash, after @command{gcore} finishes
44894 its job the program remains running without any change.
44897 @c man begin OPTIONS gcore
44900 Dump all memory mappings. The actual effect of this option depends on
44901 the Operating System. On @sc{gnu}/Linux, it will disable
44902 @code{use-coredump-filter} (@pxref{set use-coredump-filter}) and
44903 enable @code{dump-excluded-mappings} (@pxref{set
44904 dump-excluded-mappings}).
44906 @item -o @var{prefix}
44907 The optional argument @var{prefix} specifies the prefix to be used
44908 when composing the file names of the core dumps. The file name is
44909 composed as @file{@var{prefix}.@var{pid}}, where @var{pid} is the
44910 process ID of the running program being analyzed by @command{gcore}.
44911 If not specified, @var{prefix} defaults to @var{gcore}.
44915 @c man begin SEEALSO gcore
44917 The full documentation for @value{GDBN} is maintained as a Texinfo manual.
44918 If the @code{info} and @code{gdb} programs and @value{GDBN}'s Texinfo
44919 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 gdbinit GDB initialization scripts
44939 @c man begin SYNOPSIS gdbinit
44940 @ifset SYSTEM_GDBINIT
44941 @value{SYSTEM_GDBINIT}
44950 @c man begin DESCRIPTION gdbinit
44951 These files contain @value{GDBN} commands to automatically execute during
44952 @value{GDBN} startup. The lines of contents are canned sequences of commands,
44955 the @value{GDBN} manual in node @code{Sequences}
44956 -- shell command @code{info -f gdb -n Sequences}.
44962 Please read more in
44964 the @value{GDBN} manual in node @code{Startup}
44965 -- shell command @code{info -f gdb -n Startup}.
44972 @ifset SYSTEM_GDBINIT
44973 @item @value{SYSTEM_GDBINIT}
44975 @ifclear SYSTEM_GDBINIT
44976 @item (not enabled with @code{--with-system-gdbinit} during compilation)
44978 System-wide initialization file. It is executed unless user specified
44979 @value{GDBN} option @code{-nx} or @code{-n}.
44982 the @value{GDBN} manual in node @code{System-wide configuration}
44983 -- shell command @code{info -f gdb -n 'System-wide configuration'}.
44986 @ref{System-wide configuration}.
44990 User initialization file. It is executed unless user specified
44991 @value{GDBN} options @code{-nx}, @code{-n} or @code{-nh}.
44994 Initialization file for current directory. It may need to be enabled with
44995 @value{GDBN} security command @code{set auto-load local-gdbinit}.
44998 the @value{GDBN} manual in node @code{Init File in the Current Directory}
44999 -- shell command @code{info -f gdb -n 'Init File in the Current Directory'}.
45002 @ref{Init File in the Current Directory}.
45007 @c man begin SEEALSO gdbinit
45009 gdb(1), @code{info -f gdb -n Startup}
45011 The full documentation for @value{GDBN} is maintained as a Texinfo manual.
45012 If the @code{info} and @code{gdb} programs and @value{GDBN}'s Texinfo
45013 documentation are properly installed at your site, the command
45019 should give you access to the complete manual.
45021 @cite{Using GDB: A Guide to the GNU Source-Level Debugger},
45022 Richard M. Stallman and Roland H. Pesch, July 1991.
45026 @node gdb-add-index man
45027 @heading gdb-add-index
45028 @pindex gdb-add-index
45029 @anchor{gdb-add-index}
45031 @c man title gdb-add-index Add index files to speed up GDB
45033 @c man begin SYNOPSIS gdb-add-index
45034 gdb-add-index @var{filename}
45037 @c man begin DESCRIPTION gdb-add-index
45038 When @value{GDBN} finds a symbol file, it scans the symbols in the
45039 file in order to construct an internal symbol table. This lets most
45040 @value{GDBN} operations work quickly--at the cost of a delay early on.
45041 For large programs, this delay can be quite lengthy, so @value{GDBN}
45042 provides a way to build an index, which speeds up startup.
45044 To determine whether a file contains such an index, use the command
45045 @kbd{readelf -S filename}: the index is stored in a section named
45046 @code{.gdb_index}. The index file can only be produced on systems
45047 which use ELF binaries and DWARF debug information (i.e., sections
45048 named @code{.debug_*}).
45050 @command{gdb-add-index} uses @value{GDBN} and @command{objdump} found
45051 in the @env{PATH} environment variable. If you want to use different
45052 versions of these programs, you can specify them through the
45053 @env{GDB} and @env{OBJDUMP} environment variables.
45057 the @value{GDBN} manual in node @code{Index Files}
45058 -- shell command @kbd{info -f gdb -n "Index Files"}.
45065 @c man begin SEEALSO gdb-add-index
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.
45084 @node GNU Free Documentation License
45085 @appendix GNU Free Documentation License
45088 @node Concept Index
45089 @unnumbered Concept Index
45093 @node Command and Variable Index
45094 @unnumbered Command, Variable, and Function Index
45099 % I think something like @@colophon should be in texinfo. In the
45101 \long\def\colophon{\hbox to0pt{}\vfill
45102 \centerline{The body of this manual is set in}
45103 \centerline{\fontname\tenrm,}
45104 \centerline{with headings in {\bf\fontname\tenbf}}
45105 \centerline{and examples in {\tt\fontname\tentt}.}
45106 \centerline{{\it\fontname\tenit\/},}
45107 \centerline{{\bf\fontname\tenbf}, and}
45108 \centerline{{\sl\fontname\tensl\/}}
45109 \centerline{are used for emphasis.}\vfill}
45111 % Blame: doc@@cygnus.com, 1991.